# Academy Source: https://support.myapps.ai/academy Structured lessons and quizzes for the models you actually use The [Pixio Academy](https://beta.pixio.myapps.ai/academy) teaches specific models through short lessons with quizzes at the end. Where these docs explain the product, Academy teaches the craft — how to get good output from a particular model. Courses by category and model, with lessons and quizzes. ## How it's organized Courses are grouped by **category** (image, video, audio, 3D), then by **model**, then into **lessons**. Each course ends with a quiz that checks the material the lessons actually covered. ## Docs or Academy? | | These docs | Academy | | ------ | ----------------------------- | ---------------------------------- | | Answer | "How does this feature work?" | "How do I prompt this model well?" | | Shape | Reference | Lessons, in order | | Scope | The product | Specific models | Read the docs to find the button. Read Academy to get a better result once you've found it. ## A sensible path [Quickstart](/quickstart) then [Generate](/generate), so you know how the studio works. Choose a model you'll actually use rather than sampling everything. Depth in one beats a tour of ten. Work the lessons and the quiz for that model. Apply it on low-cost image generations before spending video credits. Academy courses target specific models, so the advice is concrete rather than general prompt-craft folklore. If you're about to spend real credits on a model, its course is the cheapest thing you can read first. ## Related * [Models](/models) — how to choose a model in the first place * [Generate](/generate) — where you apply it * [Prompt History](/library/prompt-history) — keep the prompts that work # Onboarding & quick-action defaults Source: https://support.myapps.ai/account/onboarding Pick the models that power your quick actions — and change them later The first time you open Pixio you're walked through a short setup that picks **which AI model runs when you use each quick action**. It's the only setting that changes how the rest of the app behaves, which is why it comes first. ## The wizard Four steps, all skippable: | Step | What it does | | -------------- | ---------------------------------------------------- | | **Welcome** | Explains what you're choosing | | **Essentials** | A default model for each core quick action | | **Optional** | Fine-tune the rest, or keep the recommended defaults | | **Review** | Confirm your choices | You can **skip this step**, **use default** on any single item, or **skip onboarding** entirely. Skipping isn't a penalty — Pixio ships sensible recommended defaults, and nothing is locked in. Nothing here is permanent. Every choice is editable later in your profile settings. ## What a quick-action default actually controls [Chained actions](/generate#chained-actions) — **Create Video**, **Upscale**, **Remove BG**, **Add Lipsync**, **Create 3D Model** and the rest — run a model without asking you which one. The default you set here is the model each of those uses. That's the practical payoff: one click does the thing with *your* preferred model instead of a generic one. ## Changing them later **Settings → Quick-action defaults** — "Choose which AI model runs when you use each quick action on the generate page." Worth revisiting when: * You've found a model you prefer for a specific job * A new model lands that's better or cheaper than your current pick * Your plan changed and better models became available * A default is costing more than you want per click Check your defaults after upgrading a plan. Onboarding picks from what your plan could reach at the time — an upgrade doesn't retroactively improve your defaults. ## Costs A quick action costs the same as running that model in [Generate](/generate). The default doesn't change the price — it changes which price you pay per click. If a chained action feels expensive, the default behind it is the thing to change. ## Related * [Generate](/generate) — where quick actions live * [Models](/models) — choosing between families * [Credits & Plans](/credits) — what things cost * [Account settings](/account/settings) # Account settings Source: https://support.myapps.ai/account/settings Profile, email, password, MFA, linked accounts, and quick-action defaults **Settings** is where you manage your profile, how you sign in, and the model defaults behind quick actions. ## What's on the page | Card | What it does | | ------------------------------- | ---------------------------------------------------- | | **Your Profile Picture** | Upload or change your avatar | | **Your Name** | Your display name | | **Update your Email** | Change the email on the account | | **Update your Password** | Set a new password | | **Multi-Factor Authentication** | Add a second factor to sign-in | | **Linked Accounts** | Social or email identities connected to this account | | **Language** | Interface language | | **Theme Color** | Appearance | | **Quick-action defaults** | Which model runs for each quick action | | **Danger Zone** | Account deletion, where enabled | ## Quick-action defaults The setting most worth revisiting: *"Choose which AI model runs when you use each quick action on the generate page."* These are the models behind [chained actions](/generate#chained-actions) — Create Video, Upscale, Remove BG, Add Lipsync and the rest. You pick them during [onboarding](/account/onboarding); change them here when a better or cheaper model turns up, or after a plan upgrade opens models you couldn't reach before. A chained action costs whatever its default model costs. If one click feels expensive, the default behind it is the thing to change — not the action. ## Multi-factor authentication Adds a second factor on top of your password. Worth enabling: your account can spend credits and holds every asset you've made. ## Linked accounts More than one sign-in method can point at the same account. If you signed up with email and later linked a social login, either gets you in. ## Password Changing your password invalidates existing sessions. That's the blunt way to revoke access an app or device still holds — see [Agents & plugins](/integrations/agents-and-plugins). Locked out? Use the reset link on the sign-in page. ## Danger Zone Personal account deletion appears here **when enabled for your deployment** — it's a configuration flag, so it may not be present. Deleting an account removes its generations and uploads. Download anything you want to keep first. Not reversible. ## What isn't here | Looking for | Go to | | ----------------------- | -------------------------------------- | | Plan, payment, invoices | [Billing](/billing/billing) | | Credit balance | [Credits & Plans](/credits) | | API keys | [Integrations](/integrations/api-keys) | | Team members and roles | [Accounts & Teams](/billing/account) | Personal and team accounts have separate settings. Team settings live on the team's own page and are managed by an owner. ## Related * [Onboarding & quick-action defaults](/account/onboarding) * [Accounts & Teams](/billing/account) * [Billing](/billing/billing) # Analytics Source: https://support.myapps.ai/advanced/analytics Your generation stats, quick actions, and recent work **Analytics** shows what you've actually been making — how much, with which models, and from which families. ## What's on the page Jump straight into the surface you use most. An interactive overview of your activity. Which model families you lean on, ranked by generation count. Your latest generations, ready to open or chain onward. ## What it's good for **Deciding your plan.** The model and company breakdown tells you where your credits actually go. Compare that against the [Maker page](https://beta.pixio.myapps.ai/maker) to see whether the models you use are included on a plan or metered — that answers the upgrade question with data instead of a guess. **Spotting waste.** Heavy volume on one model with little to show for it usually means the prompt or the reference is wrong, not the model. Cheaper to fix the input than to keep buying runs. **Finding what worked.** Recent work is the fastest route back to a result you want to build on — chain it onward in [Generate](/generate). Empty page? It fills in once you've generated. New accounts show no activity yet. ## Related * [Credits & Plans](/credits) — what things cost * [Maker Mode](/advanced/maker-mode) — which models are included versus metered * [My Assets](/assets) — the full library rather than just recent items * [Prompt History](/library/prompt-history) — every prompt you've used # CamView Source: https://support.myapps.ai/advanced/camview Block shots in 3D and use your phone as a camera **CamView** lets you set up a shot in 3D before generating it. Place objects and figures, position cameras, keyframe movement — then generate from a camera angle you actually chose rather than one you hoped a prompt would imply. ## Why it exists Prompting a specific camera angle is unreliable. "Low angle, subject to the left, camera pushing in" is a sentence a model may or may not honor. CamView makes it geometry instead of a description. ## What you work with | Piece | What it's for | | --------------------- | ---------------------------------------- | | **Objects & figures** | Blocking — what's in the scene and where | | **Mannequin** | Pose a figure to define body position | | **Cameras** | Where the shot is taken from | | **Keyframes** | Camera and figure movement over time | | **Timeline** | Sequence the keyframes | | **Shots gallery** | Shots you've set up in the scene | | **Environment** | The surrounding context | | **Templates** | Start from a prepared scene | | **Coach** | Guidance while setting up | ## Setting up a shot Add objects and figures and place them. Pose the mannequin if body position matters. Move to the view you want and set the camera there. You can also aim the camera at a selected object. Add camera keyframes for a move; keyframe figures for animation. Sequence them on the timeline. Generate from the scene. The result becomes a normal Pixio generation. ## Phone as camera (VCam) VCam turns your phone into the camera with a **live viewfinder** — move the phone and the shot moves. Physical framing instead of dragging a 3D camera with a mouse. Connect via the VCam option in CamView. VCam controller pairing and live sensor streaming are app-only — they can't be driven through the API. Scene documents themselves are scriptable. ## Saving Scenes are documents, saved as you work. * Leaving with unsaved changes prompts you first * If a scene changed elsewhere, reload before saving so you don't overwrite the newer version * **Reset** clears objects, cameras, and keyframes — your last save is kept * **Clear** removes objects and figures but keeps cameras and keyframes Reset and Clear differ in an important way: Clear preserves your camera work, Reset doesn't. If you've spent time on camera moves, use Clear. ## Credits Blocking, posing, camera work, and keyframing are free — you're arranging a scene, not generating. Credits are spent when you render. That's the whole economic argument for CamView: iterate on framing for free, then pay once for the shot you chose. ## Where it fits * **[Cinema Studio](/create/cinema-studio)** — plan the sequence of shots; VCam and keyframed cameras appear there too * **[Video Agent](/create/video-agent)** — generate a full episode * **[Generate](/generate)** — a single clip, prompt-driven # Maker Mode Source: https://support.myapps.ai/advanced/maker-mode Unlimited generations on included models, and how to tell which ones those are **Maker Mode** is the top plan — $199.99/month or $1,999.99/year. It includes **15,000 recurring credits**, **unlimited generations** on a large set of models, and **unlimited LoRAs**. ## What "unlimited" actually means This is the part worth being precise about. On Maker, models fall into two buckets: | Bucket | Behavior | | --------------------------- | --------------------------------------------------------------- | | **Included with Maker** | Using these does **not** deduct credits. Effectively unlimited. | | **Uses your Maker credits** | Available on Maker, but still spends from your balance. | So "forget about credits" holds for the included set — not for literally every model in the catalog. The expensive frontier models are the ones most likely to stay metered. Pixio publishes exactly which models are in which bucket at [beta.pixio.myapps.ai/maker](https://beta.pixio.myapps.ai/maker). Check there before planning heavy work — it's authoritative and updated as the catalog changes. ## Reading the Maker page The page separates models into: * **Included with Maker** — no credit deduction * **Uses your credits** — available, still metered * **Also included on lower plans** — included on Maker *and* at least one cheaper plan, which may mean you don't need Maker for your workload That last group is the useful one before you upgrade. If the models you actually use are included on Premium, Premium may be the better buy. ## What else you get | | | | --------------------- | ----------------------------------------- | | **Recurring credits** | 15,000, included in the subscription | | **Unlimited LoRAs** | No cap on how many custom models you keep | | **Everything below** | All Starter, Pro, and Premium features | ## Is it worth it? Check [Analytics](/advanced/analytics) — it breaks your generations down by model and company. Included, metered, or already on a lower plan? If your models are included, Maker removes a real ceiling. If they're metered, credit packs may be cheaper. Don't upgrade to Maker on the assumption that everything becomes free. If your work is concentrated in metered models, you'll pay \$199.99 and still spend credits. ## LoRA training Training requires Premium or above; Maker removes the cap on how many you keep. See [Model Training](/advanced/model-training). ## Related * [Credits & Plans](/credits) — all plans and prices * [Analytics](/advanced/analytics) — what you're actually using # Model Training Source: https://support.myapps.ai/advanced/model-training Train custom LoRA models for personalized image generation **Model Training** trains a **LoRA** — a small custom model that teaches Pixio a specific subject, character, style, or product from your own images. Once trained, you trigger it from a prompt in [Generate](/generate). LoRA training requires **Premium** or above. **Maker Mode** removes the limit on how many you keep. See [Credits & Plans](/credits). ## What you can train | Training style | Use it for | | -------------- | --------------------------------------- | | **Subject** | A specific person, animal, or object | | **Character** | A recurring character across images | | **Style** | A look or aesthetic rather than a thing | | **Product** | A product rendered consistently | | **General** | Everything else | ## Model types | Type | Notes | | --------------------- | --------------------------------------------------------------------------------------------- | | **Flux Dev** | General-purpose starting point | | **Flux Dev Portrait** | Tuned for faces and people | | **Flux Pro** | Highest quality — takes around **15 minutes**, and uses a Finetune ID rather than a Model URL | | **Flux Turbo** | Fastest; cost scales with step count | **Flux Pro training can take about 15 minutes — don't refresh the page until it completes.** ## Setting up a training run Upload individual images or a zip archive. There's a minimum and maximum image count — the form tells you if you're outside it. The **trigger word** is what you type in a prompt to invoke the LoRA. Pick something distinctive that won't collide with ordinary vocabulary. See the tables above. Portrait for faces, Style when you're teaching a look rather than a subject. More steps can mean better quality and cost more credits. Start moderate. **Face crop** for people, **Is Style** when teaching an aesthetic, a fine-tune comment, and resume-from-checkpoint to continue a previous run. The job is queued and appears under **Active Jobs**. ## Dataset quality This is where results are won or lost — far more than in the settings: * **Variety beats volume.** Different angles, lighting, and backgrounds teach the subject; twenty near-identical shots teach the background. * **Consistent subject, varied context.** The thing you're training should be the only constant. * **Crop to the subject.** Don't make the model guess what it's learning. * **Quality in, quality out.** Blurry or low-resolution inputs produce a blurry LoRA. * **For style**, use images sharing the aesthetic but *not* the subject — otherwise you train the subject by accident. A bad dataset can't be fixed with more steps. If a LoRA won't converge, rebuild the dataset rather than paying for a longer run. ## Credits Training costs credits, and cost scales with step count on Flux Turbo. If your balance is short, training won't start — you'll be told to add credits and try again. **Check your balance before starting a long run.** ## Using your LoRA Once trained, it appears in [Generate](/generate) for models that support custom weights. Include the **trigger word** in your prompt to invoke it. Starting and cancelling training jobs is app-only — you can read training *status* through the API but not start a job. See [what you can automate](/integrations/what-you-can-automate). # Run Workflow API Source: https://support.myapps.ai/aitutor-api/aitutor-api-docs POST /api/v1/run/{workflow_id} Execute a workflow with custom input parameters and receive the full generated result. ### Overview This endpoint allows you to execute a specific workflow by providing its ID and required input parameters. The response is returned after the model finishes generating — ideal for server-side integrations where you don't need real-time streaming. ### Endpoint * **HTTP Method**: POST * **URL**: `/api/v1/run/{workflow_id}` > Replace `{workflow_id}` with the actual workflow ID (e.g. `wf_abc123`). ### Features Receive the **complete generated output** in a single JSON response. Authenticate with your **API secret key** via the Authorization header. Automatically receive **source citations** when the model uses web search. Monitor your usage with **rate limit headers** in every response. ### Authentication Requires a Bearer token in the `Authorization` header. ``` Authorization: Bearer YOUR_API_KEY ``` ### Request **Path Parameters** | Name | Required | Type | Description | | ------------ | -------- | ------ | ------------------------------------------------ | | workflow\_id | Yes | string | The ID of the workflow to run (e.g. `wf_abc123`) | **Request Body** Content-Type: `application/json` Pass your workflow's input variables as key/value pairs. The keys must match the variable names defined in your workflow template. ```json theme={null} { "story": "a story about a dog named charlie" } ``` ### Example Request ```bash theme={null} curl --request POST \ --url https://aitutor-api.vercel.app/api/v1/run/wf_abc123 \ --header 'Authorization: Bearer YOUR_API_KEY' \ --header 'Content-Type: application/json' \ --data '{"story": "a story about a dog named charlie"}' ``` ### Response A successful response returns the full generated text and optional citations: ```json theme={null} { "success": true, "result": "Once upon a time, there was a friendly dog named Charlie...", "citations": ["https://example.com/source"] } ``` | Field | Type | Description | | --------- | --------- | -------------------------------------------------- | | success | boolean | Always `true` on success | | result | string | The generated text output | | citations | string\[] | Optional. Web sources if the model used web search | **Response Headers** | Header | Description | | --------------------- | ---------------------------- | | x-ratelimit-limit | Your rate limit per window | | x-ratelimit-remaining | Remaining requests in window | ### Error Responses Missing or invalid API key. ```json theme={null} { "error": "Unauthorized. Please provide a valid secret key.", "success": false } ``` No credits remaining on the account. ```json theme={null} { "error": "No credits remaining. Please add credits to continue using the service.", "success": false, "code": "invalid_billing" } ``` Workflow not found or not published. ```json theme={null} { "error": "Workflow not found", "success": false, "code": "workflow_not_found" } ``` Rate limit exceeded. ```json theme={null} { "error": "Rate limit exceeded", "success": false } ``` Something went wrong during execution. ```json theme={null} { "error": "Failed to run workflow", "success": false, "code": "internal_server_error" } ``` ### Code Examples ```python theme={null} import requests url = "https://aitutor-api.vercel.app/api/v1/run/wf_abc123" headers = { "Authorization": "Bearer YOUR_API_KEY", "Content-Type": "application/json", } data = { "story": "a story about a dog named charlie", } response = requests.post(url, headers=headers, json=data) print(response.json()) ``` ```javascript theme={null} const response = await fetch( "https://aitutor-api.vercel.app/api/v1/run/wf_abc123", { method: "POST", headers: { Authorization: "Bearer YOUR_API_KEY", "Content-Type": "application/json", }, body: JSON.stringify({ story: "a story about a dog named charlie", }), } ); const data = await response.json(); console.log(data.result); ``` ```bash theme={null} curl --request POST \ --url https://aitutor-api.vercel.app/api/v1/run/wf_abc123 \ --header 'Authorization: Bearer YOUR_API_KEY' \ --header 'Content-Type: application/json' \ --data '{"story": "a story about a dog named charlie"}' ``` ### Additional Notes * **Input Variables**: The request body keys must exactly match the variable names in your workflow template (e.g. `{{story}}`). * **Max Duration**: Requests can run for up to 300 seconds before timing out. * **Credits**: Each successful run deducts credits from your organization's balance based on token usage. * **Streaming Alternative**: If you need real-time output as it's generated, use the [Streaming Workflow](/docs/stream-workflow) endpoint instead. # AI Tutor API Info Source: https://support.myapps.ai/aitutor-api/aitutor-api-info AI Tutor API: Empowering Applications with Intelligent Tutoring Capabilities ### The AI Tutor API offers developers a powerful and user-friendly way to integrate intelligent tutoring capabilities into their applications, enabling personalized learning experiences with just a few lines of code. Launch your next AI project in hours, not months. Our platform handles hosting, security, and delivery so you can focus on building. Create workflows with any leading AI model, deploy them as API endpoints, and scale with confidence. #### Key Features Access **GPT-5, Claude Sonnet 4.5, Gemini 2.5 Pro, Grok 3, Llama, Mistral, DeepSeek** and more — all through a single unified API. Stream responses **token-by-token** as the AI generates content, fully compatible with the **Vercel AI SDK** and `useChat`. Design AI workflows with a **no-code builder** — set the model, template, instructions, and input variables, then deploy instantly. **Bearer key authentication**, single-use streaming tokens, server-derived rate limiting, and automatic credit enforcement on every request. Enable **web search** on any workflow — the AI will ground its response in real sources and return **citations** automatically. Optionally use your own **OpenAI, Anthropic, or Google** API keys for direct provider access with zero markup. #### Base URL All API requests are made to: ``` https://aitutor-api.vercel.app ``` #### Authentication API requests are authenticated using a **secret key** passed in the `Authorization` header: ``` Authorization: Bearer sk_your_secret_key ``` Secret keys are generated in your [Settings](https://aitutor-api.vercel.app/settings) dashboard. Each key is scoped to your organization and has a configurable rate limit (default: 10 requests/second). > Never expose your `sk_` key in client-side code. For browser-based streaming, use a [single-use stream token](/docs/stream-token) instead. #### API Endpoints `POST /api/v1/run/{workflow_id}` — Execute a workflow and receive the **full result** as JSON. `GET /api/v1/token` — Generate a **single-use token** for client-side streaming (1–300s TTL). `POST /api/v1/run/{workflow_id}/stream` — Execute a workflow and receive a **real-time text stream**. `POST /api/public/run/{short_id}` — Execute a published workflow **without authentication** (credits still apply). #### How It Works Follow these steps to go from zero to a live AI endpoint. In the dashboard, create a new workflow. Choose an **AI model**, write a **template** with `{{variables}}`, and optionally add **system instructions**. Use the built-in **Compose** tab to test your workflow with sample inputs. Toggle streaming on or off. When you're happy, **publish** it to make it available via API. Go to **Settings** and generate a secret key (`sk_...`). This key authenticates all your API requests and is tied to your organization's credit balance. Make a `POST` request to `/api/v1/run/{workflow_id}` with your inputs. For streaming, first get a token from `/api/v1/token`, then call the `/stream` endpoint. Integrate the response into your app — whether it's a **chatbot**, **content generator**, **learning platform**, or anything else. Scale with confidence. #### Quick Start Run your first workflow in seconds: ```bash theme={null} curl --request POST \ --url https://aitutor-api.vercel.app/api/v1/run/YOUR_WORKFLOW_ID \ --header 'Authorization: Bearer YOUR_API_KEY' \ --header 'Content-Type: application/json' \ --data '{"topic": "quantum computing"}' ``` Response: ```json theme={null} { "success": true, "result": "Quantum computing leverages quantum mechanical phenomena..." } ``` #### Rate Limits & Credits Each API key has a **configurable rate limit** (default: 10 requests per second) using a sliding window algorithm. Rate limit headers are included in every response: * `x-ratelimit-limit` — Your limit per window * `x-ratelimit-remaining` — Remaining requests If the rate limit is exceeded, the API returns **429 Too Many Requests**. Every workflow run deducts **credits** from your organization's balance based on token usage. If your balance reaches zero, API requests are blocked with a **402 Payment Required** error. Credit packs: * **Starter Pack** — 1,000 credits for \$25 * **Pro Pack** — 2,500 credits for \$50 (best value) Purchase credits in the **Settings** dashboard. You can also set a **spend limit** to cap monthly usage. Stream tokens are **single-use** and expire after their TTL (default: 60 seconds, max: 300 seconds). They are consumed on first use and cannot be reused. Generate them server-side, then pass to the client for secure browser-based streaming. #### Supported Models AI Tutor API provides access to 50+ models across all major providers: GPT-5, GPT-5.2, GPT-4.1, GPT-4o, o3, o4-mini, and more Claude Sonnet 4.5, Claude Opus 4.5, Claude Haiku 4.5 Gemini 2.5 Pro, Gemini 2.5 Flash Grok 3, Grok 3 Mini, Grok 4.1, Grok 4.1 Fast Sonar, Sonar Pro, Sonar Reasoning, Sonar Reasoning Pro Llama 3, Mistral, DeepSeek, Qwen 2.5, Command R+, and more #### Tips for Best Results Use **well-structured templates** with specific `{{variables}}` to get focused, relevant responses. The more context you provide in the system instructions, the better the output. Use **GPT-5 or Claude Sonnet 4.5** for complex reasoning tasks. Use **GPT-4.1 or Gemini 2.5 Flash** for fast, cost-effective responses. Use **Perplexity Sonar** when you need web-grounded answers with citations. For user-facing applications, **always use streaming** — it gives immediate feedback and feels dramatically faster than waiting for the full response. Always check for `success: false` in responses. Implement retries for **429** (rate limit) and **500** (server error) responses with exponential backoff. Keep `sk_` keys **server-side only**. For client-side streaming, generate a short-lived token via `/api/v1/token` and pass it to the browser. Tokens are single-use and auto-expire. Use the **Statistics** dashboard to track API calls, token usage, active models, and peak hours. Set a **spend limit** to prevent unexpected charges. #### Why AI Tutor API? * **Ship in Hours**: Go from idea to live AI endpoint without managing infrastructure, model hosting, or prompt engineering pipelines. * **50+ Models, One API**: Switch between OpenAI, Anthropic, Google, xAI, Perplexity, and open-source models without changing a line of code. * **Production-Ready Security**: Bearer authentication, single-use tokens, server-side rate limiting, and credit enforcement — all built in from day one. * **Real-time Streaming**: Deliver token-by-token responses compatible with the Vercel AI SDK, React, and any HTTP client. * **Transparent Pricing**: Pay only for what you use with simple credit packs. No hidden fees, no per-seat pricing, no minimum commitments. * **Full Control**: Visual workflow builder, customizable templates, spend limits, usage analytics, and bring-your-own-key support. # Stream Workflow API Source: https://support.myapps.ai/aitutor-api/aitutor-api-workflow-stream POST /api/v1/run/{workflow_id}/stream Execute a workflow and receive a real-time streamed response, fully compatible with Vercel AI SDKs such as useChat. ### Overview This endpoint executes a workflow and streams the response in real-time as the model generates tokens. It works seamlessly with the Vercel AI SDK and is ideal for building interactive chat interfaces and real-time tutoring experiences. ### Endpoint * **HTTP Method**: POST * **URL**: `/api/v1/run/{workflow_id}/stream?token=STREAM_TOKEN` > Replace `{workflow_id}` with your workflow ID and `STREAM_TOKEN` with a token from the [Stream Token](/docs/stream-token) endpoint. ### Features Receive and display **progressive responses** as the AI generates content — no waiting for the full result. Secure your requests with **single-use tokens** — no API key exposed on the client side. Integrates directly with `ai/react` hooks like **useChat** for reactive chat interfaces. Support for **various input formats** including text variables and image URLs. ### Authentication Uses a **stream token** passed as a query parameter (not a Bearer key). Get a token from [`GET /api/v1/token`](/docs/stream-token) first. ``` POST /api/v1/run/{workflow_id}/stream?token=pub_tok_xxx ``` ### Request **Path Parameters** | Name | Required | Type | Description | | ------------ | -------- | ------ | ------------------------------------------------ | | workflow\_id | Yes | string | The ID of the workflow to run (e.g. `wf_abc123`) | **Query Parameters** | Name | Required | Type | Description | | ----- | -------- | ------ | ---------------------------------------------- | | token | Yes | string | A single-use stream token from `/api/v1/token` | **Request Body** Content-Type: `application/json` Pass your workflow's input variables as key/value pairs. ```json theme={null} { "story": "a story about a dog named charlie" } ``` ### Implementation Guide First, obtain a single-use token from the token endpoint using your API key. ```bash theme={null} curl --request GET \ --url "https://aitutor-api.vercel.app/api/v1/token?ttl=60" \ --header "Authorization: Bearer YOUR_API_KEY" ``` Response: ```json theme={null} { "success": true, "token": "pub_tok_a1b2c3d4e5f6", "ttl": 60 } ``` Use the token to make a streaming request to the workflow endpoint: ```bash theme={null} curl --no-buffer --request POST \ --url "https://aitutor-api.vercel.app/api/v1/run/wf_abc123/stream?token=pub_tok_a1b2c3d4e5f6" \ --header 'Content-Type: application/json' \ --data '{"story": "a story about a dog named charlie"}' ``` Use `--no-buffer` so curl prints tokens as they arrive. Use this React component to handle streaming responses in your UI: ```typescript theme={null} "use client"; import { useCallback, useEffect, useState } from "react"; export default function StreamingText({ url, fallbackText, className, }: { url: string; fallbackText: string; className?: string; }) { const [loading, setLoading] = useState(false); const [result, setResult] = useState(""); const getData = useCallback(async () => { setLoading(true); const response = await fetch(url, { method: "POST", headers: { "Content-Type": "application/json" }, }); if (!response.ok) return null; const data = response.body; if (!data) { setResult(fallbackText); return; } const reader = data.getReader(); const decoder = new TextDecoder(); let done = false; setLoading(false); while (!done) { const { value, done: doneReading } = await reader.read(); done = doneReading; const chunkValue = decoder.decode(value); setResult((prev) => (prev ?? "") + chunkValue); } }, [url, fallbackText]); useEffect(() => { if (url) { getData(); } }, []); return loading ? (
Loading...
) : (

{result}

); } ```
For chat-style interfaces, integrate with the Vercel AI SDK's `useChat` hook: ```typescript theme={null} "use client"; import { useChat } from "ai/react"; export default function ChatPage({ token }: { token: string }) { const { messages, input, handleInputChange, handleSubmit } = useChat({ api: `https://aitutor-api.vercel.app/api/v1/run/wf_abc123/stream?token=${token}`, keepLastMessageOnError: true, }); return ( <> {messages.map((message) => (
{message.role === "user" ? "User: " : "AI: "} {message.content}
))}
); } ```
### Response The response is a real-time `text/plain` stream. Text chunks arrive as they are generated: ``` Content-Type: text/plain; charset=utf-8 ``` ``` Once upon a time, there was a friendly dog named Charlie... ``` ### Error Responses Missing, invalid, or already-used token. ```json theme={null} { "error": "Unauthorized. Please provide a valid secret key.", "success": false } ``` No credits remaining on the account. ```json theme={null} { "error": "No credits remaining. Please add credits to continue using the service.", "success": false, "code": "invalid_billing" } ``` Workflow not found or not published. ```json theme={null} { "error": "Workflow not found", "success": false } ``` Something went wrong during execution. ```json theme={null} { "error": "Failed to run workflow", "success": false, "code": "internal_server_error" } ``` ### Code Examples ```python theme={null} import requests API_KEY = "YOUR_API_KEY" WORKFLOW_ID = "wf_abc123" BASE_URL = "https://aitutor-api.vercel.app" # Step 1: Get stream token token_res = requests.get( f"{BASE_URL}/api/v1/token?ttl=60", headers={"Authorization": f"Bearer {API_KEY}"}, ) token = token_res.json()["token"] # Step 2: Stream the response response = requests.post( f"{BASE_URL}/api/v1/run/{WORKFLOW_ID}/stream?token={token}", headers={"Content-Type": "application/json"}, json={"story": "a story about a dog named charlie"}, stream=True, ) for chunk in response.iter_content(chunk_size=None, decode_unicode=True): print(chunk, end="", flush=True) ``` ```javascript theme={null} const API_KEY = "YOUR_API_KEY"; const WORKFLOW_ID = "wf_abc123"; const BASE_URL = "https://aitutor-api.vercel.app"; // Step 1: Get stream token const tokenRes = await fetch(`${BASE_URL}/api/v1/token?ttl=60`, { headers: { Authorization: `Bearer ${API_KEY}` }, }); const { token } = await tokenRes.json(); // Step 2: Stream the response const response = await fetch( `${BASE_URL}/api/v1/run/${WORKFLOW_ID}/stream?token=${token}`, { method: "POST", headers: { "Content-Type": "application/json" }, body: JSON.stringify({ story: "a story about a dog named charlie" }), } ); const reader = response.body.getReader(); const decoder = new TextDecoder(); while (true) { const { value, done } = await reader.read(); if (done) break; process.stdout.write(decoder.decode(value)); } ``` ```bash theme={null} # Step 1: Get token curl --request GET \ --url "https://aitutor-api.vercel.app/api/v1/token?ttl=60" \ --header "Authorization: Bearer YOUR_API_KEY" # Step 2: Stream (replace TOKEN with the token from step 1) curl --no-buffer --request POST \ --url "https://aitutor-api.vercel.app/api/v1/run/wf_abc123/stream?token=TOKEN" \ --header 'Content-Type: application/json' \ --data '{"story": "a story about a dog named charlie"}' ``` ### Best Practices Implement **robust error handling** for network issues, expired tokens, and invalid responses. Always check `response.ok` before reading the stream. Tokens are **single-use** — generate a fresh token before each stream request. Never cache or reuse tokens. Display appropriate **loading indicators** while waiting for the initial response chunk to arrive. Optimize your application to handle **continuous data streams** efficiently. Use `ReadableStream` readers and avoid buffering the entire response in memory. ### Additional Notes * **Single-Use Token**: The stream token is consumed on the first request — if it's expired or already used, you'll get a 401. * **Max Duration**: Streaming requests can run for up to 300 seconds before timing out. * **Credits**: The workflow owner's credits are charged based on total tokens generated. * **Non-Streaming Alternative**: Use the [`POST /api/v1/run/{workflow_id}`](/docs/run-workflow) endpoint if you don't need real-time output. # API Credit Balance Source: https://support.myapps.ai/aitutor-api/get-balance GET /api/v1/balance} Execute a workflow with custom input parameters and receive the full generated result. # Balance API Check your account credit balance using your API key. ## Endpoint **GET** `/api/v1/balance` ## Authentication Send your API key in the `Authorization` header: ``` Authorization: Bearer YOUR_API_KEY ``` ## Example Request ```bash theme={null} curl -X GET "https://aitutor-api.vercel.app/api/v1/balance" \ -H "Authorization: Bearer sk_xxxxxxxxxxxxxxxxxxxxxxxx" ``` ## Success Response (200) ```json theme={null} { "success": true, "credits": 1234 } ``` | Field | Type | Description | | ------- | ------- | ------------------------ | | success | boolean | Always `true` on success | | credits | number | Current credit balance | ## Error Responses **401 Unauthorized** — Missing or invalid API key. ```json theme={null} { "error": "Unauthorized. Please provide a valid secret key.", "success": false } ``` **429 Too Many Requests** — Rate limit exceeded. ```json theme={null} { "error": "Rate limit exceeded", "success": false } ``` **500 Internal Server Error** — Server error. ```json theme={null} { "error": "Failed to retrieve balance, please try again or contact support.", "success": false, "code": "internal_server_error" } ``` ## Response Headers | Header | Description | | --------------------- | ---------------------------- | | x-ratelimit-limit | Rate limit per window | | x-ratelimit-remaining | Remaining requests in window | # Stream Token API Source: https://support.myapps.ai/aitutor-api/get-token-workflow GET /api/v1/token Generate a secure, single-use token for authenticating streaming workflow requests. ### Overview This endpoint generates a short-lived, single-use token for the streaming endpoint. Tokens are stored securely in Redis and automatically expire after their TTL period. This is **step 1** of the two-step streaming flow. ### Endpoint * **HTTP Method**: GET * **URL**: `/api/v1/token` ### Features Each token can only be used **once** — it is deleted from Redis after the first stream request. Set a custom **time-to-live** between 1 and 300 seconds (default: 60s). Tokens are generated under your **API key** — credits are charged to your organization. Server-derived rate limiting ensures **fair usage** across all consumers. ### Authentication Requires a Bearer token in the `Authorization` header. ``` Authorization: Bearer YOUR_API_KEY ``` ### Request **Query Parameters** | Name | Required | Type | Default | Description | | ---- | -------- | ------ | ------- | --------------------------------- | | ttl | No | number | 60 | Token lifetime in seconds (1–300) | No body content is required for this GET request. ### Example Request ```bash theme={null} curl --request GET \ --url "https://aitutor-api.vercel.app/api/v1/token?ttl=60" \ --header "Authorization: Bearer YOUR_API_KEY" ``` With a custom TTL of 5 minutes: ```bash theme={null} curl --request GET \ --url "https://aitutor-api.vercel.app/api/v1/token?ttl=300" \ --header "Authorization: Bearer YOUR_API_KEY" ``` ### Response A successful response returns the token and its TTL: ```json theme={null} { "success": true, "token": "pub_tok_a1b2c3d4e5f6", "ttl": 60 } ``` | Field | Type | Description | | ------- | ------- | --------------------------------------------------- | | success | boolean | Always `true` on success | | token | string | The single-use stream token (format: `pub_tok_...`) | | ttl | number | Token lifetime in seconds | **Response Headers** | Header | Description | | --------------------- | ---------------------------- | | x-ratelimit-limit | Your rate limit per window | | x-ratelimit-remaining | Remaining requests in window | ### Error Responses Missing or invalid API key. ```json theme={null} { "error": "Unauthorized. Please provide a valid secret key.", "success": false } ``` No credits remaining on the account. ```json theme={null} { "error": "No credits remaining. Please add credits to continue using the service.", "success": false, "code": "invalid_billing" } ``` Rate limit exceeded. ```json theme={null} { "error": "Rate limit exceeded", "success": false } ``` Token generation failed. ```json theme={null} { "error": "Failed to create token, please try again or contact support.", "success": false, "code": "internal_server_error" } ``` ### Code Examples ```python theme={null} import requests url = "https://aitutor-api.vercel.app/api/v1/token?ttl=60" headers = { "Authorization": "Bearer YOUR_API_KEY", } response = requests.get(url, headers=headers) token = response.json()["token"] print(token) ``` ```javascript theme={null} const response = await fetch( "https://aitutor-api.vercel.app/api/v1/token?ttl=60", { headers: { Authorization: "Bearer YOUR_API_KEY", }, } ); const { token } = await response.json(); console.log(token); ``` ```bash theme={null} curl --request GET \ --url "https://aitutor-api.vercel.app/api/v1/token?ttl=60" \ --header "Authorization: Bearer YOUR_API_KEY" ``` ### Additional Notes * **Single-Use**: The token is deleted from Redis after its first use in a stream request. * **Auto-Expiry**: Tokens expire automatically after the TTL period — unused tokens do not count against your API quota. * **Next Step**: Use this token as a query parameter on the [Stream Workflow](/docs/stream-workflow) endpoint to start receiving real-time responses. # List Files API Source: https://support.myapps.ai/aitutor-api/list-files-rag GET /user_files View all files uploaded to your account with pagination support. ### Request Body **Content-Type**: `application/json` ```json theme={null} { "pagination": { "limit": 10, // Optional: Number of files to fetch (default: 10) "offset": 0 // Optional: Offset for pagination (default: 0) }, "order_dir": "desc" // Optional: Sort direction - "desc" or "asc" } ``` ### Authentication Before making requests to the API, you need to obtain your API key from your dashboard. ### Example Request ```json theme={null} const res = await fetch("https:/https://rag-api-llm.up.railway.app/user_files", { method: "POST", headers: { "Content-Type": "application/json", "authorization": apiKey, }, body: JSON.stringify({ pagination: { limit: 10, offset: 0 }, order_dir: "desc" }), }); ``` ### Response ```json theme={null} { "success": true, "results": [ { "type": "text", "file_id": "939b2cd7-56d5-4c44-b491-14bfc202e9a0", "created_at": "2025-01-04T03:22:10.713451+00:00", "file_name": "test content 4.txt", "team_id": "2ce522b2-2045-43ad-ab0a-2f48a9a041e9" }, { "type": "pdf", "file_id": "4db6b5d7-b199-4bfd-9914-6991cb4ba3b6", "created_at": "2025-01-04T02:56:10.252537+00:00", "file_name": "1227-2024_WhisperIt-Invoice-Taishi - Invoice.pdf", "team_id": "2ce522b2-2045-43ad-ab0a-2f48a9a041e9" } ], "pagination": { "offset": 0, "limit": 10 }, "count": 143 } ``` ### Parameters | Name | Required | Type | Description | | ---------- | -------- | ------ | ---------------------------------------- | | pagination | No | object | Pagination parameters (limit and offset) | | order\_dir | No | string | Sort direction ("desc" or "asc") | | Name | Required | Type | Description | | ------ | -------- | ------- | ----------------------------------------- | | limit | No | integer | Number of files to fetch (default: 10) | | offset | No | integer | Starting position of results (default: 0) | ### Error Responses ```json theme={null} { "success": false, "error": { "message": "Invalid API key provided", "code": "unauthorized" } } ``` ```json theme={null} { "success": false, "error": { "message": "Invalid pagination parameters", "code": "invalid_request" } } ``` # Multi-File RAG API Source: https://support.myapps.ai/aitutor-api/multi-file-rag POST /embeddings Search for embeddings relevant to a query across multiple uploaded files simultaneously. ### Request Body **Content-Type**: `application/json` ```json theme={null} { "query": "string", // Query to search across multiple files "file_ids": ["string", "string"], // Array of multiple file UUIDs to search in "k": 3 // Optional: Number of related chunks to return (default: 3) } ``` ### Authentication Before making requests to the AI Tutor RAG API, you need to obtain your API key from your dashboard. Example Request ```json theme={null} const res = await fetch("https://rag-api-llm.up.railway.app/embeddings", { method: "POST", headers: { "Content-Type": "application/json", "authorization": apiKey, }, body: JSON.stringify({ query: "tell me about ross and nice art", file_ids: [ "387dffa2-5f6c-4ad0-8d27-fb0ea0cbff83", "38b5d591-77c7-4612-a88c-ee1eaebba7bf" ] }), }); ``` ### Response ```json theme={null} { "success": true, "documents": [ { "content": "Ross created a beautiful painting of mountain landscapes.", "file_id": "387dffa2-5f6c-4ad0-8d27-fb0ea0cbff83", "score": "0.89" }, { "content": "The art gallery featured nice contemporary pieces.", "file_id": "38b5d591-77c7-4612-a88c-ee1eaebba7bf", "score": "0.76" } ] } ``` ### Error Responses ```json theme={null} { "success": false, "error": { "message": "Invalid API key provided", "code": "unauthorized" } } ``` ```json theme={null} { "success": false, "error": { "message": "No file IDs provided", "code": "invalid_request" } } ``` ```json theme={null} { "success": false, "error": { "message": "One or more file IDs are invalid", "code": "invalid_file_ids" } } ``` # RAG Search API Source: https://support.myapps.ai/aitutor-api/rag-api POST /run Search for embeddings relevant to a query in a list of uploaded files. ### Request Body **Content-Type**: `application/json` ```json theme={null} { "query": "string", // Query for which to get related chunks and embeddings "file_ids": ["string"], // Array of file UUIDs to search in "k": 3 // Optional: Number of related chunks to return (default: 3) } ``` ### Authentication Before making requests to the AI Tutor RAG API, you need to obtain your API key from your dashboard. ### Example Request ```javascript theme={null} const res = await fetch("https://rag-api-llm.up.railway.app/embeddings", { method: "POST", headers: { "Content-Type": "application/json", "authorization": apiKey, }, body: JSON.stringify({ query: "search query", file_ids: ["939b2cd7-56d5-4c44-b491-14bfc202e9a0"] }), }); ``` ### Response ```json theme={null} { "success": true, "documents": [ { "content": "my name is jon.", "file_id": "939b2cd7-56d5-4c44-b491-14bfc202e9a0", "score": "0.273" } ] } ``` ### Request Parameters | Name | Required | Type | Description | | --------- | -------- | ------- | ------------------------------------------------------------------- | | query | Yes | string | Query for which to get related chunks and embeddings | | file\_ids | Yes | array | Array of file UUIDs to search in. Must contain at least one file ID | | k | No | integer | Number of related chunks to return. Default: 3 | ### Error Responses ```json theme={null} { "success": false, "error": { "message": "Invalid API key provided", "code": "unauthorized" } } ``` ```json theme={null} { "success": false, "error": { "message": "No file IDs provided", "code": "invalid_request" } } ``` # AI Tutor RAG Info Source: https://support.myapps.ai/aitutor-api/rag-info Build powerful RAG applications ### AI Tutor RAG AI Tutor RAG is the open-source alternative to Carbon.ai. Build powerful RAG applications with any data source, at any scale. Use the AI Tutor RAG API to generate embeddings from your documents and connect them to AI. ### Getting started To get started, create an API key in the dashboard and read about how to make requests for the resources you need to access using our HTTP APIs. ### What is RAG? RAG (Retrieval-Augmented Generation) is a powerful approach that enhances AI responses with relevant information from your documents. Instead of relying solely on an AI's general knowledge, RAG searches through your specific documents to provide accurate, context-aware answers. Think of RAG as giving your AI assistant access to your organization's knowledge base. When you ask a question, it first finds the most relevant information from your documents, then uses this specific context to generate a precise answer. ### Why RAG Matters Accurate Responses: Get answers based on your actual documents, not just general AI knowledge Up-to-date Information: Your AI always references your latest documents Data Privacy: Keep your sensitive information secure while leveraging AI capabilities Cost Efficiency: Optimize AI usage by focusing on relevant context Introducing Supavec Supavec is an open-source RAG infrastructure that makes it simple to build AI applications powered by your documents. Whether you're creating a customer support bot, internal knowledge base, or document analysis tool, Supavec provides the foundation you need. # File Upload API Source: https://support.myapps.ai/aitutor-api/upload-file-rag POST /upload Upload PDF or text files to be processed and chunked for embedding search. ### Request **Content-Type**: `multipart/form-data` The POST request should be a multipart form request containing the file to be uploaded. ### Query Parameters | Name | Type | Required | Description | | -------------- | ------- | -------- | ----------------------------------------------------------------- | | chunk\_size | integer | No | The chunk size (in tokens) applied when splitting the document | | chunk\_overlap | integer | No | The chunk overlap (in tokens) applied when splitting the document | ### Authentication Before making requests to the AI Tutor RAG API, you need to obtain your API key from your dashboard. ### Example Request ```javascript theme={null} const formData = new FormData(); formData.append("file", files[0]); // Accepts PDF (.pdf) or text (.txt) files const response = await fetch("https://rag-api-llm.up.railway.app/upload_file", { method: "POST", headers: { "authorization": apiKey, }, body: formData, }); ``` Response ```javascript theme={null} { "success": true, "message": "PDF file processed successfully", "file_id": "4db6b5d7-b199-4bfd-9914-6991cb4ba3b6", "file_name": "elon_musk.txt", "chunks": 2, "chunk_size": 1000, "chunk_overlap": 200 } ``` File Requirements Requirement Description File Types PDF (.pdf) or text (.txt) files only Maximum Size 10MB Download CSV Error Responses ```javascript theme={null} { "success": false, "error": { "message": "Invalid file type. Only PDF and TXT files are supported", "code": "invalid_file_type" } } ``` ```javascript theme={null} { "success": false, "error": { "message": "File size exceeds maximum limit of 10MB", "code": "file_too_large" } } ``` ```javascript theme={null} { "success": false, "error": { "message": "Invalid API key provided", "code": "unauthorized" } } ``` # Adding Models Source: https://support.myapps.ai/aitutor/adding-models Adding Models in Plus # Adding Models ### Comprehensive Guide to Incorporating Models into Plus Expanding the capabilities of Plus with additional models not only enhances the versatility of your projects but also personalizes your work to meet specific needs. This detailed guide walks you through the process of adding models to Plus, ensuring you can leverage the full power of AI-driven functionalities with ease. ### Initial Steps: Accessing the Models Menu * Locate the Models Dropdown: Start your journey on the Plus interface by directing your attention to the top bar. Here, nestled among various options, you will find a dropdown menu situated next to the persona dropdown---a gateway to extending your Plus capabilities. ![](https://docs.myapps.ai/storage/page_media/f4f7bab0-5a52-4218-8b2d-4f7dd4f97789.png) * Selecting "Models": Click on this dropdown menu and choose the "Models" option. This action triggers the appearance of a dedicated menu, designed to manage and integrate new models into your Plus environment. ![](https://docs.myapps.ai/storage/page_media/3bb7140f-6a00-4b88-bb66-b662f4b1050d.png) ### Process of Adding New Models 1. Initiating Model Addition: Within the models menu, your attention should be drawn to the "Add" button. This is your first step towards diversifying your Plus toolkit. Clicking on this button unveils a dropdown filled with a plethora of models, each offering unique enhancements to Plus. ![](https://docs.myapps.ai/storage/page_media/007230b2-05e3-4648-84f8-72ab69f392dd.png) 2. Choosing the Perfect Model: Delve into the dropdown menu to explore the wide array of models at your disposal. From data analysis enhancements to creative content generation tools, each model is designed to cater to various aspects of your projects. Take your time to understand the functionalities of each model, ensuring it aligns with the goals of your current or future projects. 3. Incorporating an API Key: For certain models, integration with Plus may require an API key. This key acts as a secure link between Plus and the model, facilitating a smooth and efficient addition process. If your chosen model necessitates an API key, a prompt will appear for you to input this crucial piece of information. 4. Completing the Addition Process: After selecting your model and providing the necessary API key, solidify your choice by clicking on the blue "Models" button. This final step officially adds the model to the models dropdown menu, integrating it into your Plus toolkit and making it ready for use in enhancing your projects. ### Wrapping Up: Utilizing Your New Models With your selected models now added to Plus, you're poised to explore new horizons in your projects. These models not only offer specialized functionalities to tackle specific tasks but also open up opportunities for innovation and creativity. Whether you're streamlining data processing, generating unique content, or exploring novel AI applications, the models you've integrated into Plus are ready to elevate your work to new levels of efficiency and ingenuity. ### Conclusion Adding models to Plus is a straightforward yet impactful process that significantly broadens the scope of what you can achieve with this powerful platform. By carefully selecting models that resonate with your project's needs and successfully integrating them into Plus, you unlock a world of possibilities, tailor-made to enhance your productivity and creativity. Dive into the diverse world of models available in Plus, and transform the way you approach your projects with these AI-driven tools at your disposal. # Custom Prompt Tips Source: https://support.myapps.ai/aitutor/custom-prompt-tips Custom Prompt Tips # Custom Prompt Tips ### Introduction: Creating an effective prompt for AI Plus involves a strategic approach that enhances the interaction between the user and the AI. This process is crucial in achieving tailored responses that meet specific needs and preferences. Here, we'll explore a step-by-step guide on how to craft an optimized prompt using the persona creator, setting clear rules, and defining desired output qualities. Creating the Persona: 1. Start with the Persona Creator: Begin by accessing the persona creator tool. This feature allows you to describe the desired characteristics and capabilities of your AI persona. Provide a concise description focusing on what you expect from the AI in terms of behavior, expertise, and interaction style. 2. Generate and Refine the Persona: Once you submit your description, the tool will generate a detailed persona profile. Review this information carefully to ensure it aligns with your objectives. Feel free to keep useful traits and discard any elements that do not serve your purpose, keeping the persona focused and relevant to your needs. Setting the Rules: 1. Establish Clear Guidelines: Define a set of rules that the AI must follow. These rules are critical in guiding the AI's behavior and responses. For example, if accuracy and non-speculation are crucial, include a rule such as "never guess information." 2. Contextual Relevance: Tailor the rules to fit the specific task at hand. If the AI is tasked with creating educational content, emphasize the importance of clarity, factual accuracy, and engagement in the rules. Defining Output Qualities: 1. Desired Length and Detail: Decide whether you prefer concise or detailed responses. This will influence how the AI structures its replies and how much information it includes in each interaction. 2. Tone and Format: Specify the tone (formal, casual, persuasive, etc.) and the format (list, essay, bullet points) of the AI's responses. This ensures the outputs are not only informative but also presented in a manner that resonates with your audience. 3. Feedback Mechanism: Consider how the AI should handle requests for further information or clarification. This might include asking follow-up questions or providing additional resources. Crafting the Prompt: With the persona defined, rules set, and output qualities specified, you can now construct your prompt. Integrate all three aspects to guide the AI effectively. Start with a clear introduction of the task, followed by the rules, and conclude with specific instructions on the desired output format and tone. Conclusion: Optimizing your AI Plus prompt is a dynamic process that involves a clear understanding of your goals, the creation of a tailored persona, and precise instructions on how the AI should operate and respond. By meticulously setting up the persona, establishing rules, and defining output characteristics, you can enhance the efficiency and relevance of your AI interactions, leading to more meaningful and productive outcomes. This structured approach not only streamlines communication but also maximizes the potential of AI to cater to specific needs and preferences. # Import Export Options Source: https://support.myapps.ai/aitutor/importing-exporting-chats Importing and Exporting Chats # Importing & Exporting Chats ### Introduction: Managing your interactions effectively in AI platforms involves the ability to export and import chat data. This capability allows users to keep records of their conversations, analyze data, or migrate information between different systems or accounts. Here's a straightforward guide on how to handle exporting and importing chats in Plus, ensuring you maintain control over your data. Exporting Chats: 1. Access the Export Feature: Navigate to the bottom of the left side menu in Plus and locate the 'Export' button. 2. Select Export Options: * Current Chat: If you wish to export the chat you are currently viewing, select this option. It allows you to download the ongoing conversation without including any others. * All Chats: Choose this if you want to export every chat you have engaged in. This is useful for a comprehensive backup or when you need a full record of all interactions. 3. Choose the Format: Decide on the format in which you would like to download your chats. Common formats include PDF, CSV, or TXT files. Select the one that best suits your needs for review or archiving. Importing Chats: 1. Access the Import Feature: Right next to the export button, you will find the 'Import' button. Click on this to start the process of importing your chat data. 2. Select Import Options: * JSON File: If you have chat data saved in a JSON format, you can upload this file. This option is particularly useful for maintaining the structure and details of your chats. * CPT Shared Link: If you have a shared link to a chat or a project, you can import it directly using this option. This method is handy when transferring data from another user or system. 3. Using JSON for AI Tutor: For those transferring chats in AI Tutor within the same account, it is recommended to use JSON files for exporting and importing. This ensures compatibility and seamless integration of your chats. Conclusion: Exporting and importing chats in Plus is a straightforward process that enhances your data management capabilities. Whether you are looking to back up important conversations, analyze interactions, or migrate chats to a new account, understanding how to use these features effectively is crucial. By following the outlined steps for both exporting and importing, you can ensure that your chat data is handled securely and efficiently, allowing you to make the most out of your AI platform experience. # Input and Output Options Source: https://support.myapps.ai/aitutor/input-output-options Input and Output Options # Input and Output Options AI Tutor provides a suite of comprehensive tools designed to simplify and enhance your chat generation experience. Understanding that information can come in various formats, AI Tutor allows you to upload files, including PDFs, enabling the AI to read and integrate the content into your interactions. Additionally, the platform supports the uploading of pictures, further enriching the context for your AI conversations. Beyond the standard chat functionality, AI Tutor introduces several innovative options to tailor your interaction according to your needs. These include: 1. Write: This option focuses on generating text-based outputs. It's ideal for creating detailed responses, essays, reports, or any written content, leveraging the AI's capability to understand and articulate ideas clearly. 2. Draw: For those who require visual representations, the Draw feature enables the AI to create sketches or diagrams. This is particularly useful in fields like design, architecture, or any area where visual aids can enhance understanding. 3. Reason + Act: This advanced feature allows the AI to not only process information but also make inferences and suggest actions based on the provided data. It's especially useful for scenario-based learning, problem-solving, and decision-making exercises. 4. Browse Command: This advanced feature allows the AI to scrape a website and keep the content in context. Use /browse to activate! Each option modifies how the AI processes your input and generates output, allowing for a customized chat generation experience tailored to your specific project or learning objective. Whether you're looking to draft a document, visualize concepts, or engage in complex reasoning, AI Tutor's versatile tools are designed to accommodate a wide range of educational and creative needs. # Persona Generator Source: https://support.myapps.ai/aitutor/persona-generator Persona Generator # Persona Generator The Persona Creator stands out as a powerful tool designed to elevate the quality and specificity of your prompts. Ideal for users who seek to develop more intricate and detailed prompts, this feature provides a versatile platform where both video and text inputs are welcome. By allowing you to add multimedia elements or detailed textual information, the Persona Creator enriches the context and depth of the persona you wish to create. Once you input your desired information, the tool works efficiently to analyze and compile a comprehensive persona profile in just a few minutes. This process not only saves time but also ensures that the resulting persona is rich in detail and closely aligned with your initial vision. The detailed persona generated can then be seamlessly integrated into the creation of your custom prompt. This step is crucial for those looking to achieve a higher degree of personalization in their AI interactions. Whether you're aiming to simulate specific scenarios, embody particular knowledge bases, or capture a unique tone of voice, the Persona Creator equips you with the means to do so with precision and creativity. By leveraging the Persona Creator, you're not just crafting a prompt; you're sculpting a distinct AI personality that can significantly enhance your learning experience, creative projects, or any other endeavor that benefits from tailored AI assistance. # Personalities Source: https://support.myapps.ai/aitutor/personalities Adding Personalities. Welcome to AI Tutor Plus, your comprehensive guide to unlocking the full potential of artificial intelligence in your learning and development journey. Our platform is designed to cater to a wide range of needs, offering a diverse selection of AI chatbots, each boasting its unique "personality" and area of expertise. Choosing the right AI for your situation is straightforward on our platform. Simply identify your current need or challenge, and select the AI personality that best aligns with your objectives. Each AI is designed to offer in-depth assistance, enabling you to achieve your goals with efficiency and ease. We understand that our existing AI personalities might not always match your specific needs, we provide a custom personality feature. This enables you to create an AI tailored exactly to your requirements, ensuring a perfect fit for your unique needs. Creating your own custom AI personality is a straightforward and engaging process. Here's how you can get started: 1. Navigate to the Custom Icon: Within the AI Tutor Plus platform, look for the "Custom" icon. This is your gateway to crafting a personalized AI experience. 2. Craft Your Prompt: Once you've clicked on the custom icon, you'll be prompted to describe the personality and capabilities you wish your AI to possess. This is a critical step, as the more detailed and specific you are, the better the AI can be tailored to meet your needs. Consider including information about the subject matter, the tone of interaction (e.g., formal, casual, humorous), and any specific knowledge areas or skills you want the AI to have. 3. Refine and Review: After submitting your initial prompt, you may have the opportunity to refine and adjust the details. This iterative process ensures that the final custom AI personality truly reflects your vision and objectives. By leveraging the custom personality feature, you can go beyond the pre-defined AI options and create a learning companion that's as unique as your learning journey itself. Whether you need an AI that combines elements from multiple disciplines, one that speaks in a particular style, or anything in between, the power to customize ensures that your educational or developmental needs are met with precision and creativity. Embark on a personalized learning adventure with AI Tutor Plus and discover the limitless possibilities of custom AI personalities tailored just for you. # Account credit balance Source: https://support.myapps.ai/api-reference/account-credit-balance /api/pixio-api.json get /credits # Agent guide (Markdown, or format=json) Source: https://support.myapps.ai/api-reference/agent-guide-markdown-or-format=json /api/pixio-api.json get /guide # Apply validated canvas operations. Source: https://support.myapps.ai/api-reference/apply-validated-canvas-operations /api/pixio-project-api.json post /api/v1/canvas/{id}/operations # Apply validated spatial-board operations. Source: https://support.myapps.ai/api-reference/apply-validated-spatial-board-operations /api/pixio-project-api.json post /api/v1/boards/{id}/operations # Apply validated timeline operations with the editor engine. Source: https://support.myapps.ai/api-reference/apply-validated-timeline-operations-with-the-editor-engine /api/pixio-project-api.json post /api/v1/editor/projects/{id}/operations # Batch signed attachment download URLs Source: https://support.myapps.ai/api-reference/batch-signed-attachment-download-urls /api/pixio-api.json get /assets/download # Bulk-delete assets you own Source: https://support.myapps.ai/api-reference/bulk-delete-assets-you-own /api/pixio-api.json delete /assets # Clean public URL for an image (upload file or mirror URL) Source: https://support.myapps.ai/api-reference/clean-public-url-for-an-image-upload-file-or-mirror-url /api/pixio-api.json post /images # Clean public URL for ANY media (image video audio) Source: https://support.myapps.ai/api-reference/clean-public-url-for-any-media-image-video-audio /api/pixio-api.json post /media # Create a board project. Source: https://support.myapps.ai/api-reference/create-a-board-project /api/pixio-project-api.json post /api/v1/boards # Create a Cam View scene project. Source: https://support.myapps.ai/api-reference/create-a-cam-view-scene-project /api/pixio-project-api.json post /api/v1/cam-view/scenes # Create a canvas project. Source: https://support.myapps.ai/api-reference/create-a-canvas-project /api/pixio-project-api.json post /api/v1/canvas # Create a chat conversation project. Source: https://support.myapps.ai/api-reference/create-a-chat-conversation-project /api/pixio-project-api.json post /api/v1/chat/conversations # Create a editor project project. Source: https://support.myapps.ai/api-reference/create-a-editor-project-project /api/pixio-project-api.json post /api/v1/editor/projects # Create a storyboard project. Source: https://support.myapps.ai/api-reference/create-a-storyboard-project /api/pixio-project-api.json post /api/v1/cinema/storyboards # Create a structured Video Agent outline from a prompt. Source: https://support.myapps.ai/api-reference/create-a-structured-video-agent-outline-from-a-prompt /api/pixio-project-api.json post /api/v1/video-agent/projects/from-prompt # Create a Video Agent project project. Source: https://support.myapps.ai/api-reference/create-a-video-agent-project-project /api/pixio-project-api.json post /api/v1/video-agent/projects # Create or replace a locked character. Source: https://support.myapps.ai/api-reference/create-or-replace-a-locked-character /api/pixio-project-api.json post /api/v1/characters # Credit ledger entries (top-ups, debits, refunds) Source: https://support.myapps.ai/api-reference/credit-ledger-entries-top-ups-debits-refunds /api/pixio-api.json get /credits/ledger # Delete a generation you own Source: https://support.myapps.ai/api-reference/delete-a-generation-you-own /api/pixio-api.json delete /generations/{id} # Delete an asset you own Source: https://support.myapps.ai/api-reference/delete-an-asset-you-own /api/pixio-api.json delete /assets/{id} # Delete one board project. Source: https://support.myapps.ai/api-reference/delete-one-board-project /api/pixio-project-api.json delete /api/v1/boards/{id} # Delete one Cam View scene project. Source: https://support.myapps.ai/api-reference/delete-one-cam-view-scene-project /api/pixio-project-api.json delete /api/v1/cam-view/scenes/{id} # Delete one canvas project. Source: https://support.myapps.ai/api-reference/delete-one-canvas-project /api/pixio-project-api.json delete /api/v1/canvas/{id} # Delete one chat conversation project. Source: https://support.myapps.ai/api-reference/delete-one-chat-conversation-project /api/pixio-project-api.json delete /api/v1/chat/conversations/{id} # Delete one editor project project. Source: https://support.myapps.ai/api-reference/delete-one-editor-project-project /api/pixio-project-api.json delete /api/v1/editor/projects/{id} # Delete one locked character. Source: https://support.myapps.ai/api-reference/delete-one-locked-character /api/pixio-project-api.json delete /api/v1/characters/{name} # Delete one storyboard project. Source: https://support.myapps.ai/api-reference/delete-one-storyboard-project /api/pixio-project-api.json delete /api/v1/cinema/storyboards/{id} # Delete one Video Agent project project. Source: https://support.myapps.ai/api-reference/delete-one-video-agent-project-project /api/pixio-project-api.json delete /api/v1/video-agent/projects/{id} # Direct and save a Cinema storyboard from a prompt. Source: https://support.myapps.ai/api-reference/direct-and-save-a-cinema-storyboard-from-a-prompt /api/pixio-project-api.json post /api/v1/cinema/storyboards/from-prompt # Direct and save a keyed Cam View scene from a prompt. Source: https://support.myapps.ai/api-reference/direct-and-save-a-keyed-cam-view-scene-from-a-prompt /api/pixio-project-api.json post /api/v1/cam-view/scenes/from-prompt # Direct and save a layered canvas from a prompt. Source: https://support.myapps.ai/api-reference/direct-and-save-a-layered-canvas-from-a-prompt /api/pixio-project-api.json post /api/v1/canvas/from-prompt # Direct and save a spatial board from a prompt. Source: https://support.myapps.ai/api-reference/direct-and-save-a-spatial-board-from-a-prompt /api/pixio-project-api.json post /api/v1/boards/from-prompt # Discover supported public API surfaces and contracts. Source: https://support.myapps.ai/api-reference/discover-supported-public-api-surfaces-and-contracts /api/pixio-project-api.json get /api/v1/capabilities # Get API Key with Code Source: https://support.myapps.ai/api-reference/endpoint/auth GET /auth-response/ ### Description This endpoint retrieves an API key using a request ID, specifically built for ComfyUI workflow uploads. ### Parameters * **request\_id** (required, path): The request ID used to retrieve the API key. ### Responses #### 200 OK `{ "api_key": "string", // The returned API key "name": "string" // Name associated with the API key }` #### 201 API Key Not Ready json Copy code `{ "message": "string" // Message indicating the API key is not yet ready }` #### 401 Invalid or Expired Token text Copy code `Invalid or expired token` #### 500 Error Fetching API Key json Copy code `{ "error": "string" // Error message detailing why the key retrieval failed }` # Run a Workflow Source: https://support.myapps.ai/api-reference/endpoint/create POST /run ### Description This endpoint allows you to run a workflow by providing a deployment ID and optional input parameters. ### Request Body **Content-Type**: `application/json` ```json theme={null} { "deployment_id": "string", // The ID of the deployment to run (required) "inputs": { // Optional inputs for the workflow "key": "string or number" } } ``` # Run a Workflow Source: https://support.myapps.ai/api-reference/endpoint/delete POST /run ### Description This endpoint allows you to run a workflow by providing a deployment ID and optional input parameters. ### Request Body **Content-Type**: `application/json` ```json theme={null} { "deployment_id": "string", // The ID of the deployment to run (required) "inputs": { // Optional inputs for the workflow "key": "string or number" } }` ``` ### Responses #### 200 OK ```json theme={null} `{ "run_id": "string" // The ID of the run that has been queued }` ``` #### 401 Invalid or Expired Token ```json theme={null} `Invalid or expired token` ``` #### 500 Error Creating Run ```json theme={null} `{ "error": "string" // Error message detailing why the run creation failed }` ``` # Get Workflow Run Output Source: https://support.myapps.ai/api-reference/endpoint/get GET /run ### Description This endpoint retrieves the output of a workflow run by providing the `run_id`. ### Parameters * **run\_id** (required, query): The ID of the run for which the output is being retrieved. ### Responses #### 200 OK ```json theme={null} `{ "id": "string", // Run ID "workflow_version_id": "string", // Workflow version ID "workflow_inputs": { // Inputs provided to the workflow "input_text": "string", "input_image": "https://somestatic.png" }, "workflow_id": "string", // ID of the workflow "machine_id": "string", // Machine used to run the workflow "origin": "manual | api | public-share", // Origin of the workflow run "status": "not-started | running | uploading | success | failed", // Current status "ended_at": "string", // End timestamp "created_at": "string", // Creation timestamp "started_at": "string" // Start timestamp }` ``` #### 400 Workflow Not Found ```json theme={null} `{ "code": "400", "message": "Workflow not found" }` ``` #### 401 Invalid or Expired Token ```json theme={null} `Invalid or expired token` ``` #### 500 Error Getting Output ```json theme={null} `{ "error": "string" // Error message detailing why the output retrieval failed }` ``` # Upload Files to Storage Source: https://support.myapps.ai/api-reference/endpoint/upload GET /upload-url ### Description This endpoint generates an upload URL for file uploads, such as images, to the storage. ### Parameters * **type** (required, query): The MIME type of the file (e.g., `image/png`, `image/jpg`, `image/jpeg`). * **file\_size** (required, query): The size of the file to be uploaded. ### Responses #### 200 OK ```json theme={null} `{ "upload_url": "string", // URL to upload the file "file_id": "string", // File ID for the uploaded file "download_url": "string" // Download URL for the file }` ``` #### 401 Invalid or Expired Token ```json theme={null} `Invalid or expired token` ``` #### 500 Error Generating Upload URL ```json theme={null} `{ "error": "string" // Error message detailing why the URL generation failed }` ``` # Estimate credit cost before generating Source: https://support.myapps.ai/api-reference/estimate-credit-cost-before-generating /api/pixio-api.json post /generations/estimate # Generate media Source: https://support.myapps.ai/api-reference/generate-media /api/pixio-api.json post /generate # Get a saved workflow and its node definition. Source: https://support.myapps.ai/api-reference/get-a-saved-workflow-and-its-node-definition /api/pixio-project-api.json get /api/v1/workflows/{id} # Get one asset with a signed URL Source: https://support.myapps.ai/api-reference/get-one-asset-with-a-signed-url /api/pixio-api.json get /assets/{id} # Get one board project. Source: https://support.myapps.ai/api-reference/get-one-board-project /api/pixio-project-api.json get /api/v1/boards/{id} # Get one Cam View scene project. Source: https://support.myapps.ai/api-reference/get-one-cam-view-scene-project /api/pixio-project-api.json get /api/v1/cam-view/scenes/{id} # Get one canvas project. Source: https://support.myapps.ai/api-reference/get-one-canvas-project /api/pixio-project-api.json get /api/v1/canvas/{id} # Get one chat conversation project. Source: https://support.myapps.ai/api-reference/get-one-chat-conversation-project /api/pixio-project-api.json get /api/v1/chat/conversations/{id} # Get one editor project project. Source: https://support.myapps.ai/api-reference/get-one-editor-project-project /api/pixio-project-api.json get /api/v1/editor/projects/{id} # Get one locked character. Source: https://support.myapps.ai/api-reference/get-one-locked-character /api/pixio-project-api.json get /api/v1/characters/{name} # Get one model-training job. Source: https://support.myapps.ai/api-reference/get-one-model-training-job /api/pixio-project-api.json get /api/v1/training/{id} # Get one prompt and its saved generation params. Source: https://support.myapps.ai/api-reference/get-one-prompt-and-its-saved-generation-params /api/pixio-project-api.json get /api/v1/prompt-library/{id} # Get one storyboard project. Source: https://support.myapps.ai/api-reference/get-one-storyboard-project /api/pixio-project-api.json get /api/v1/cinema/storyboards/{id} # Get one Video Agent project project. Source: https://support.myapps.ai/api-reference/get-one-video-agent-project-project /api/pixio-project-api.json get /api/v1/video-agent/projects/{id} # Inputs for one model Source: https://support.myapps.ai/api-reference/inputs-for-one-model /api/pixio-api.json get /params # List a workflow’s runs Source: https://support.myapps.ai/api-reference/list-a-workflow’s-runs /api/pixio-api.json get /workflows/{id}/runs # List board projects. Source: https://support.myapps.ai/api-reference/list-board-projects /api/pixio-project-api.json get /api/v1/boards # List Cam View scene projects. Source: https://support.myapps.ai/api-reference/list-cam-view-scene-projects /api/pixio-project-api.json get /api/v1/cam-view/scenes # List canvas projects. Source: https://support.myapps.ai/api-reference/list-canvas-projects /api/pixio-project-api.json get /api/v1/canvas # List chat conversation projects. Source: https://support.myapps.ai/api-reference/list-chat-conversation-projects /api/pixio-project-api.json get /api/v1/chat/conversations # List editor project projects. Source: https://support.myapps.ai/api-reference/list-editor-project-projects /api/pixio-project-api.json get /api/v1/editor/projects # List locked reusable characters. Source: https://support.myapps.ai/api-reference/list-locked-reusable-characters /api/pixio-project-api.json get /api/v1/characters # List model-training jobs and outputs. Source: https://support.myapps.ai/api-reference/list-model-training-jobs-and-outputs /api/pixio-project-api.json get /api/v1/training # List models Source: https://support.myapps.ai/api-reference/list-models /api/pixio-api.json get /models # List prompts from generation history. Source: https://support.myapps.ai/api-reference/list-prompts-from-generation-history /api/pixio-project-api.json get /api/v1/prompt-library # List saved workflows Source: https://support.myapps.ai/api-reference/list-saved-workflows /api/pixio-api.json get /workflows # List storyboard projects. Source: https://support.myapps.ai/api-reference/list-storyboard-projects /api/pixio-project-api.json get /api/v1/cinema/storyboards # List Video Agent project projects. Source: https://support.myapps.ai/api-reference/list-video-agent-project-projects /api/pixio-project-api.json get /api/v1/video-agent/projects # List your assets (uploads + generated), newest first Source: https://support.myapps.ai/api-reference/list-your-assets-uploads-+-generated-newest-first /api/pixio-api.json get /assets # List your generations (all statuses), newest first Source: https://support.myapps.ai/api-reference/list-your-generations-all-statuses-newest-first /api/pixio-api.json get /generations # Model detail + inputs Source: https://support.myapps.ai/api-reference/model-detail-+-inputs /api/pixio-api.json get /models/{id} # OpenAPI 3.1 schema for the public API Source: https://support.myapps.ai/api-reference/openapi-31-schema-for-the-public-api /api/pixio-api.json get /openapi.json # Optimize a prompt for a media type Source: https://support.myapps.ai/api-reference/optimize-a-prompt-for-a-media-type /api/pixio-api.json post /prompts/optimize # Plan, credit quota, and API concurrency limit Source: https://support.myapps.ai/api-reference/plan-credit-quota-and-api-concurrency-limit /api/pixio-api.json get /subscription # Poll a generation Source: https://support.myapps.ai/api-reference/poll-a-generation /api/pixio-api.json get /generations/{id} # Poll a workflow run Source: https://support.myapps.ai/api-reference/poll-a-workflow-run /api/pixio-api.json get /workflows/{id}/runs/{runId} # Read the additive project API OpenAPI contract. Source: https://support.myapps.ai/api-reference/read-the-additive-project-api-openapi-contract /api/pixio-project-api.json get /api/v1/platform/openapi.json # Rename an asset (uploads only) Source: https://support.myapps.ai/api-reference/rename-an-asset-uploads-only /api/pixio-api.json patch /assets/{id} # Run a workflow Source: https://support.myapps.ai/api-reference/run-a-workflow /api/pixio-api.json post /workflows/{id}/runs # Signed attachment download URL for an asset Source: https://support.myapps.ai/api-reference/signed-attachment-download-url-for-an-asset /api/pixio-api.json get /assets/{id}/download # Update one board project. Source: https://support.myapps.ai/api-reference/update-one-board-project /api/pixio-project-api.json patch /api/v1/boards/{id} # Update one Cam View scene project. Source: https://support.myapps.ai/api-reference/update-one-cam-view-scene-project /api/pixio-project-api.json patch /api/v1/cam-view/scenes/{id} # Update one canvas project. Source: https://support.myapps.ai/api-reference/update-one-canvas-project /api/pixio-project-api.json patch /api/v1/canvas/{id} # Update one chat conversation project. Source: https://support.myapps.ai/api-reference/update-one-chat-conversation-project /api/pixio-project-api.json patch /api/v1/chat/conversations/{id} # Update one editor project project. Source: https://support.myapps.ai/api-reference/update-one-editor-project-project /api/pixio-project-api.json patch /api/v1/editor/projects/{id} # Update one locked character. Source: https://support.myapps.ai/api-reference/update-one-locked-character /api/pixio-project-api.json patch /api/v1/characters/{name} # Update one storyboard project. Source: https://support.myapps.ai/api-reference/update-one-storyboard-project /api/pixio-project-api.json patch /api/v1/cinema/storyboards/{id} # Update one Video Agent project project. Source: https://support.myapps.ai/api-reference/update-one-video-agent-project-project /api/pixio-project-api.json patch /api/v1/video-agent/projects/{id} # Upload media into your assets (alias of uploads) Source: https://support.myapps.ai/api-reference/upload-media-into-your-assets-alias-of-uploads /api/pixio-api.json post /assets # Upload media to user assets (stored + signed URLs) Source: https://support.myapps.ai/api-reference/upload-media-to-user-assets-stored-+-signed-urls /api/pixio-api.json post /uploads # My Assets Source: https://support.myapps.ai/assets Everything you've generated and everything you've uploaded **My Assets** is your library. Two tabs: **Generated** and **Uploads**. ## Generated Every generation on your account — images, video, audio, 3D — with its type, status, and when it was made. The same data as the history in [Generate](/generate), in a dedicated, paginated list. Per item you can: * **Open** it in the preview * **Download** it * **Publish** it to the public gallery, or unpublish later * **Delete** it ## Uploads Files you've uploaded — references, source images, audio tracks. Anything uploaded through a generation form, Boards, Canvas, or the editor lands here. Uploads are reusable. Once a file is here you can select it as an input on any model that accepts that type, without uploading it again. ## Why you'll come back here Feed a past generation or upload into a new model — the basis of chaining. Delete failed experiments and things you don't need. Send your best work to the public gallery. Download the final files. Faster than downloading and re-uploading: from the [Generate](/generate) history, hover a result and use a **chained action** to push it into the next model directly. ## Scope Assets belong to the account that created them. On a **team account**, every member sees the same shared assets. On a personal account they're yours alone. See [Teams & Account](/billing/account). Deleting is permanent. Download anything you want to keep before removing it. ## Related * [Prompt History](/library/prompt-history) — the prompts behind these outputs * [Boards](/create/boards) — organize work into projects * [Share links](/library/share-links) — share without a Pixio account * [what you can automate](/integrations/what-you-can-automate) — assets are fully scriptable # Accounts & Teams Source: https://support.myapps.ai/billing/account Personal accounts, team accounts, members, and roles Pixio has two kinds of account. Understanding which you're in explains where your credits and assets went. | | Personal account | Team account | | ------- | ---------------- | -------------------------- | | Credits | Yours alone | **One shared pool** | | Assets | Private to you | Shared with every member | | Billing | You | Usually an owner | | Members | Just you | Invited members with roles | You can belong to both — a personal account plus one or more teams. Switch between them in the app; whichever you're in is the one being billed. Generating in the wrong context spends the wrong balance and puts the output in the wrong library. Check which account you're in before starting expensive work. ## Team accounts A team account is a shared workspace: one plan, one credit pool, one asset library, many people. Good for a studio or agency where several people generate against a single budget and need each other's outputs. Not necessary if you just want your own work in one place. ## Members and roles Two roles, with **owner** above **member**: | Role | Can | | ---------- | ----------------------------------------------------------------------------------- | | **Owner** | Everything a member can, plus manage members, invites, roles, billing, and settings | | **Member** | Work in the team — generate, use shared assets, spend the shared credits | Permissions cover roles, billing, settings, members, and invites — all owner-level. Any member can spend the team's credits. There's no per-member credit limit, so a shared pool needs a shared understanding of who's generating what. ## Inviting people Available to owners. Invite by email with the role they should have. The invite is joined from the link. Invites can **expire** — an expired one needs re-sending. Members can be removed, and roles changed, by an owner. ## Billing Personal and team accounts bill separately, each with its own plan and credit balance. A team's billing is managed on the team's own [Billing](/billing/billing) page, normally by an owner. Upgrading your personal plan does nothing for your team, and vice versa. They're independent subscriptions. ## Ownership transfer Team ownership can be transferred to another member — the path to take when the current owner leaves. Do it *before* the account they own is deactivated. ## Related * [Billing](/billing/billing) — plans and payment * [Credits & Plans](/credits) — what things cost * [My Assets](/assets) — shared on a team account # Billing Source: https://support.myapps.ai/billing/billing Why your statement says AI TUTOR APP, and where to get billing help # Credit card statement info for our products When you subscribe to **AI Tutor** or **Pixio**, you may notice that your billing statement displays **AI TUTOR APP** as the transaction label. This is a shared billing identifier for both services, ensuring a streamlined and efficient payment process. ## Why Does It Say AI TUTOR APP? Both **AI Tutor** and **Pixio** are part of the same ecosystem and use a unified payment processor. To keep things simple, we use the label **AI TUTOR APP** for all transactions associated with these products. ## Example Statement Here's how the charge might appear on your credit card statement: AI TUTOR APP - \$XX.XX The amount charged will correspond to the subscription tier or services you've selected for either **AI Tutor** or **Pixio**. ## Need Help with Billing? If you have any questions or concerns about your billing statement, feel free to reach out to our support team. We’re happy to clarify any charges and ensure your experience is seamless. * **Email:** [rcohen@mytsi.org](mailto:rcohen@mytsi.org) * **Support Hours:** Monday to Friday, 9:00 AM - 5:00 PM (EST) Thank you for using **AI Tutor** and **Pixio**! We appreciate your trust in our products. # Custom nodes & models Source: https://support.myapps.ai/comfyui/custom-nodes-and-models What your workflow can rely on in the ComfyUI environment A workflow only runs if everything it references exists on the machine executing it — every custom node, checkpoint, LoRA, and VAE. This is the most common reason a graph that works on your desktop fails as a deployment. Node and model availability is a property of the **environment**, not of your workflow. Deploying a version captures your graph — it does not bundle the nodes or weights the graph depends on. ## Check before you build The reliable source of truth is the environment itself, not documentation: Browse the node list. If a custom node isn't there, your graph can't use it. Open the checkpoint, LoRA, or VAE dropdowns in your graph. What's listed is what's installed. A manual run surfaces a missing dependency immediately, and far more clearly than a failed API run will. ## When something you need is missing Ask before redesigning. Getting a node or checkpoint added is often easier than rebuilding a graph to avoid it — and a rebuild that works around a missing dependency usually costs quality. If you can't get it added, the fallbacks in order of preference: 1. **Substitute a node** that's already installed for the same operation 2. **Move that step out of ComfyUI** — do it with a catalog model through the [Pixio API](/integrations/overview) and pass the result in 3. **Split the workflow** so the unavailable step is handled elsewhere in your pipeline ## Version your dependencies deliberately Because dependencies live in the environment rather than your graph, they can change underneath a deployment that you haven't touched. A deployment that has worked for months can start failing without you changing anything, if a node or checkpoint it depends on changes in the environment. When an untouched deployment breaks, compare the run's `machine_id` against a previously successful run before assuming your graph is at fault. Practical habits: * **Note what each deployment depends on** — the custom nodes and weights it needs. Future-you debugging a failure will want that list. * **Keep graphs as lean as possible.** Every extra custom node is another thing that can go missing. * **Test after any environment change** rather than finding out through a production failure. * **Prefer widely-used nodes** over obscure ones. They're likelier to stay installed. ## Diagnosing a dependency failure A missing node or model shows up as a failed run, not as a clear "not installed" message. To confirm that's the cause: | Check | What it tells you | | ------------------------------- | ----------------------------------------------------------- | | Run the same graph **manually** | Fails there too → dependency or graph problem, not the API | | The run's `workflow_version_id` | Confirms which revision actually executed | | The run's `machine_id` | Two machines behaving differently points at the environment | | Remove the suspect node | If it then runs, you've found it | More in [Troubleshooting](/comfyui/troubleshooting). ## Related * [Workflows & deployments](/comfyui/workflows-and-deployments) — why deploying, not saving, is what changes behavior * [Running a workflow](/comfyui/running-a-workflow) — statuses and polling * [Overview](/comfyui/overview) — how Pixio ComfyUI differs from the catalog API # Files & uploads Source: https://support.myapps.ai/comfyui/files-and-uploads Get images and other media into a workflow run Workflows that take an image, video, or audio input need that file reachable by URL. You either point at a URL you already host, or upload the file to Pixio storage first. ## Option 1 — a URL you already have If the file is already publicly reachable, pass the URL straight through in `inputs`: ```json theme={null} { "deployment_id": "your-deployment-id", "inputs": { "input_image": "https://example.com/reference.png" } } ``` Simplest path — but the URL must be publicly reachable for the whole run. Signed URLs that expire mid-run, or links behind auth, will fail. ## Option 2 — upload to Pixio storage For local files, get an upload URL first. `GET /upload-url` takes: | Parameter | Required | Description | | ----------- | -------- | -------------------------------------------------- | | `type` | Yes | MIME type — `image/png`, `image/jpg`, `image/jpeg` | | `file_size` | Yes | Size of the file in bytes | Response: ```json theme={null} { "upload_url": "...", "file_id": "...", "download_url": "..." } ``` Three-step flow: Send the MIME type and exact byte size. Upload the bytes directly to the returned URL. Pass `download_url` as your workflow's image input. ```js theme={null} const fs = require('fs'); const buffer = fs.readFileSync('reference.png'); const { upload_url, download_url } = await client.getUploadUrl({ type: 'image/png', file_size: buffer.length, }); await fetch(upload_url, { method: 'PUT', headers: { 'Content-Type': 'image/png' }, body: buffer, }); await client.runWorkflow({ deployment_id: 'your-deployment-id', inputs: { input_image: download_url }, }); ``` `file_size` must be the **actual** byte length. A mismatch between the size you declare and the bytes you send causes the upload to be rejected. Use `buffer.length` or `fs.statSync(path).size` — never an estimate. ## Getting it right * **Match the MIME type to the file.** A `.jpg` sent as `image/png` may fail or confuse the graph. * **Reuse uploads.** One `download_url` can feed many runs — don't re-upload the same reference per run. * **Upload before you run.** The file must exist before the graph reaches the node that reads it. * **Keep the `file_id`.** It's your handle on the stored file. ## Outputs Outputs are produced during the run's `uploading` phase — which is exactly why `uploading` isn't terminal. Read output URLs only after the status is `success`; earlier, they may be missing or incomplete. Treat output URLs as something to fetch and store on your side if you need them long-term. Don't assume a URL from a run last month still resolves. ## Errors | Response | Meaning | | ------------------------------ | --------------------------------------------------- | | `401 Invalid or expired token` | Check the Bearer header | | `500` with an `error` field | Upload URL generation failed — the message says why | # Getting started Source: https://support.myapps.ai/comfyui/getting-started Install the SDK, authenticate, and run your first workflow **Prerequisites** — Node.js 18.10.0 or higher, an API token from the Pixio dashboard, and a deployed workflow with a `deployment_id`. ## 1. Install the SDK ```bash theme={null} npm install @pixio/api-sdk ``` ```bash theme={null} yarn add @pixio/api-sdk ``` ## 2. Store your key Put the token in a `.env` file — never in source control. ```bash theme={null} PIXIO_API_KEY=your-api-key-here ``` Add `.env` to `.gitignore` before your first commit. A leaked token can run workflows on your account. ## 3. Create the client ```js theme={null} const PixioAPI = require('@pixio/api-sdk'); const client = new PixioAPI(process.env.PIXIO_API_KEY); client.getProjects().then((projects) => { console.log(projects); }); ``` If that returns your projects, your token works. ## 4. Run a workflow Pass the `deployment_id` and whatever inputs your graph expects: ```js theme={null} client .runWorkflow({ deployment_id: 'your-deployment-id', inputs: { input_text: 'Example input', input_image: 'https://example.com/image.png', }, }) .then((response) => { console.log('Run ID:', response.run_id); }); ``` The call returns as soon as the run is **queued**. It does not wait for the result. ## 5. Retrieve the output Use the `run_id` to fetch the run: ```js theme={null} client.getWorkflowOutput('your-run-id').then((output) => { console.log('Output:', output); }); ``` The response includes a `status`. Keep polling until it's `success`: ```js theme={null} async function waitForRun(runId, { intervalMs = 2000, timeoutMs = 600000 } = {}) { const deadline = Date.now() + timeoutMs; while (Date.now() < deadline) { const run = await client.getWorkflowOutput(runId); if (run.status === 'success') return run; if (run.status === 'failed') throw new Error(`Run ${runId} failed`); await new Promise((resolve) => setTimeout(resolve, intervalMs)); } throw new Error(`Run ${runId} timed out`); } ``` Wait for `success`, not merely for `running` to end. A run passes through `uploading` after execution finishes but before outputs are readable. ## Keeping the SDK current ```bash theme={null} npm update @pixio/api-sdk ``` ## Local development notes If you're running a local server alongside the SDK and hit a port clash: ```js theme={null} const PORT = process.env.PORT || 3333; app.listen(PORT, () => console.log(`Server running on port ${PORT}`)); ``` An `EADDRINUSE` error means something else already holds that port — change the port or stop the other process. ## Next How versions and deployments work, and how to ship changes safely. Feed images and other media into a run. Inputs, polling, and output handling in depth. Token errors, failed runs, and stuck jobs. # Pixio ComfyUI Source: https://support.myapps.ai/comfyui/overview Run your own ComfyUI workflows in the cloud over an API **Pixio ComfyUI** is a separate product from the Pixio app. You build a ComfyUI workflow, deploy it, and queue runs over HTTP — your graph, your custom nodes, your models, on managed GPUs. This is not [Pixio Integrations](/integrations/overview). That API gives you 500+ curated models with fixed inputs. **This** runs *your* ComfyUI graph. Different product, different API, different key. ## Which one do I want? | | Pixio app & API | Pixio ComfyUI | | ------------ | ----------------------------- | --------------------------------- | | What runs | Curated catalog models | Your own ComfyUI workflow | | Control | The inputs each model exposes | Every node in your graph | | Custom nodes | No | Yes | | GPU choice | Managed | **You can override per run** | | Setup | None | Build and deploy a workflow first | If a catalog model does the job, use [the app](/generate) — it's faster and there's nothing to maintain. Reach for ComfyUI when you need control the catalog can't give you. ## Base URL and auth ``` https://pixio-api-workers-production.up.railway.app/api ``` Every request sends an API key as a Bearer token: ``` Authorization: Bearer YOUR_API_KEY ``` See [API keys](/comfyui/api-keys). ## Endpoints | | Endpoint | Purpose | | ------ | ----------------------- | ------------------------------------- | | `POST` | `/run/deployment/queue` | Queue a run | | `GET` | `/run/{run_id}` | Get a run's status and outputs | | `POST` | `/run/{run_id}/cancel` | Cancel a run | | `POST` | *your endpoint* | Receive run status updates by webhook | ## How it works Author your graph in ComfyUI, exposing inputs with Pixio's external input nodes — see [workflow inputs](/comfyui/files-and-uploads). A deployment makes a specific workflow version callable. It gets a `deployment_id`, which is what your code references. `POST /run/deployment/queue` with the `deployment_id` and your inputs. You get a `run_id` back immediately — execution is asynchronous. Pass a `webhook` when queueing and receive status pushes including the terminal one, or poll `GET /run/{run_id}`. Webhooks are preferred. ## Core concepts | Term | What it is | | -------------------- | ------------------------------------------------------- | | **Workflow** | Your ComfyUI graph | | **Workflow version** | A specific saved revision | | **Deployment** | A version made callable — identified by `deployment_id` | | **Run** | One execution, identified by `run_id` | | **Machine** | The environment the run executes on | The distinction that matters: you deploy a **version** and you call a **deployment**. Editing your graph changes nothing about your API calls until you deploy — which is what makes production safe. ## Billing and concurrency * **Billing posts at the terminal state**, for the time actually used * **Queue time is unbilled** * Your plan's **Concurrent GPUs** limit decides how many run at once; runs beyond it wait in `queued` rather than failing That last point is why a burst of queued runs isn't an error. They're waiting for a GPU, not broken. ## Where to next Queue your first run and fetch its outputs. All nine statuses, webhooks vs polling, cancelling. Exposing inputs with external nodes, including images. Creating, scoping, and revoking keys. # Running a workflow Source: https://support.myapps.ai/comfyui/running-a-workflow Trigger a run, poll it, and read the output Running is two calls: start it, then poll it. There's no synchronous "run and return the image" — execution is always asynchronous. ## Start a run `POST /run` with the deployment and its inputs: ```json theme={null} { "deployment_id": "your-deployment-id", "inputs": { "input_text": "a lighthouse at dusk", "input_image": "https://example.com/reference.png" } } ``` Only `deployment_id` is required — `inputs` depends on what your graph expects. The response is immediate: ```json theme={null} { "run_id": "..." } ``` That means **queued**, not finished. Store the `run_id`. ## Poll the run `GET /run` with the `run_id`: ```json theme={null} { "id": "...", "workflow_id": "...", "workflow_version_id": "...", "machine_id": "...", "workflow_inputs": { "input_text": "a lighthouse at dusk", "input_image": "https://somestatic.png" }, "origin": "api", "status": "success", "created_at": "...", "started_at": "...", "ended_at": "..." } ``` The run echoes back the inputs it actually received and the exact version that ran — invaluable when a result is wrong and you need to know whether the input or the graph was at fault. ## Statuses | Status | Terminal? | What to do | | ------------- | --------- | -------------------------------------------------- | | `not-started` | No | Queued — keep polling | | `running` | No | Executing — keep polling | | `uploading` | **No** | Finished executing, storing outputs — keep polling | | `success` | Yes | Read the output | | `failed` | Yes | Stop; inspect the error | `uploading` is the one that trips people up. It means execution finished but outputs aren't ready. Only `success` and `failed` are terminal — anything else means poll again. ## Polling well ```js theme={null} async function waitForRun(client, runId, { intervalMs = 2000, timeoutMs = 600000, } = {}) { const deadline = Date.now() + timeoutMs; while (Date.now() < deadline) { const run = await client.getWorkflowOutput(runId); if (run.status === 'success') return run; if (run.status === 'failed') throw new Error(`Run ${runId} failed`); await new Promise((resolve) => setTimeout(resolve, intervalMs)); } throw new Error(`Run ${runId} timed out after ${timeoutMs}ms`); } ``` Guidelines that save trouble: * **Poll every 1–3 seconds.** Tighter than that is wasted requests; looser makes fast runs feel slow. * **Always set a timeout.** A run that never reaches a terminal state must not hang your process forever. * **Treat `failed` as expected**, not exceptional. Graphs fail on bad inputs; handle it. * **Persist the `run_id`** before you start polling. If your process restarts mid-run, that ID is the only way back to the result. * **Back off on errors.** A `500` while polling doesn't mean the run failed — retry the poll. For long runs, don't hold an HTTP request open while polling. Store the `run_id`, return to your caller, and poll from a background job. Video-length graphs outlast most request timeouts. ## Errors | Response | Meaning | | ------------------------------ | ------------------------------------------------------------ | | `401 Invalid or expired token` | Bad, expired, or missing token — check the Bearer header | | `400 Workflow not found` | The `deployment_id` or `run_id` doesn't exist or isn't yours | | `500` with an `error` field | Server-side failure; the message says why | More detail in [Troubleshooting](/comfyui/troubleshooting). ## Concurrency Nothing forces you to run one at a time — start several and poll each `run_id` independently. Runs execute on machines, so throughput depends on available capacity rather than on your polling loop. If you're firing a large batch, keep a bounded number in flight rather than launching hundreds at once. # Troubleshooting Source: https://support.myapps.ai/comfyui/troubleshooting Token errors, failed runs, stuck jobs, and changes that don't take effect ## Authentication Three usual causes, in order of likelihood: 1. **The header is malformed.** It must be exactly `Authorization: Bearer ` — one space, no quotes around the token. 2. **The token isn't loading.** Log whether your env var is defined (never its value). A missing `.env` in production is the classic version of this. 3. **The token was regenerated.** Creating a new one doesn't keep the old one alive. Update every place that used it. Almost always the environment variable. `.env` files are typically not deployed — set the variable in your host's configuration, then confirm the deployed process actually sees it. `201` from `GET /auth-response/{request_id}` means **the key isn't ready yet**, not that something failed. Keep polling. ## Runs The `deployment_id` or `run_id` doesn't exist, or doesn't belong to your account. Check for a copy-paste truncation, and confirm you're using a **deployment** ID rather than a workflow ID — they're different things. See [workflows & deployments](/comfyui/workflows-and-deployments). Failures are usually the graph, not the platform. Check, in order: * **Inputs** — does `workflow_inputs` on the run match what you meant to send? The run echoes back what it actually received. * **Reachability** — was every input URL publicly reachable for the whole run? Expired signed URLs are a common culprit. * **The graph by hand** — run it manually in ComfyUI with the same inputs. If it fails there too, it's the graph. Confirm you're treating `uploading` as non-terminal — a poll loop that only waits for `running` to end will appear to hang. Check `started_at`: if it's null, the run is still queued and waiting on capacity, not stuck. You read them too early. Outputs are written during `uploading`. Only read after `success`. Compare `workflow_version_id` across the two runs — if they differ, a deployment changed underneath you. If they match, compare `machine_id`. Both are recorded on every run for exactly this. ## Changes not taking effect **You didn't deploy it.** Saving a workflow in ComfyUI doesn't change what your `deployment_id` points at. Deploy the new version — see [workflows & deployments](/comfyui/workflows-and-deployments). Input names are a contract. Renaming one breaks every existing caller. Either restore the old name or update all callers — and prefer deploying breaking changes as a new deployment. ## Uploads `file_size` must be the exact byte length of what you send. Use `buffer.length` or `fs.statSync(path).size`, not an estimate. Also confirm the MIME type matches the actual file. Pass `download_url` as the input — not `upload_url` (write-only) and not `file_id`. ## Local development Another process holds that port. Stop it, or run on a different one: ```js theme={null} const PORT = process.env.PORT || 3333; app.listen(PORT, () => console.log(`Server running on port ${PORT}`)); ``` Reinstall cleanly — delete `node_modules` and your lockfile, then `npm install`. If it persists, check the installed SDK version against the latest. ## Reporting a problem Include these and the answer comes back faster: * The **`run_id`** — it carries the version, machine, inputs, and timing * What you **expected** versus what happened * Whether the same graph works **manually** in ComfyUI * The **status** the run reached, and its `started_at` / `ended_at` Never include your API token in a bug report, a screenshot, or a support message. # Workflows & deployments Source: https://support.myapps.ai/comfyui/workflows-and-deployments Versions, deployments, and shipping graph changes without breaking production Four ideas, and the relationship between them is the whole model: ``` Workflow → Workflow version → Deployment → Run (the graph) (a saved revision) (callable) (one execution) ``` | Term | Identified by | What it is | | -------------------- | --------------------- | ----------------------------- | | **Workflow** | `workflow_id` | Your ComfyUI graph | | **Workflow version** | `workflow_version_id` | A specific saved revision | | **Deployment** | `deployment_id` | A version exposed to the API | | **Run** | `run_id` | One execution of a deployment | ## Why versions matter **Your code calls a deployment, not a workflow.** Editing your graph in ComfyUI changes nothing about what your API calls do until you deploy the new version. That's the safety property worth internalizing: * Experiment freely — production keeps running the deployed version * Ship deliberately — deploying is the moment behavior changes * Roll back by pointing at the previous version The corollary catches people out: **saving your workflow does not update your API.** If a fix isn't showing up in runs, check that you deployed it. ## Inputs are a contract Your graph's input nodes define what callers must send. The `inputs` object in a run request maps to those names: ```json theme={null} { "deployment_id": "your-deployment-id", "inputs": { "input_text": "a lighthouse at dusk", "input_image": "https://example.com/reference.png" } } ``` Renaming or removing an input node is a **breaking change** for every caller. Treat input names as a public API: * **Adding** an optional input is safe * **Renaming** one breaks existing callers * **Removing** one breaks existing callers * **Changing the type** breaks them quietly, which is worse Need to change inputs incompatibly? Deploy it as a new deployment and migrate callers over, rather than mutating the one in production. ## Inspecting a version `GET /workflow-version/{version_id}` returns a version's detail — useful when a run behaved unexpectedly and you want to confirm which revision actually executed. Every run records its `workflow_version_id`, so you can always trace a result back to the exact graph that produced it. ## Machines Runs execute on a **machine**, recorded as `machine_id` on the run. Two runs of the same deployment on different machines should behave identically — if they don't, the machine is where to look, and the `machine_id` is what to report. ## Run origin Every run records where it came from: | Origin | Means | | -------------- | ---------------------------- | | `manual` | Triggered by hand | | `api` | Triggered through the API | | `public-share` | Triggered via a public share | Useful for separating your own testing from real traffic when you're reviewing history. ## Practical workflow Get the graph right in ComfyUI first. Debugging a graph through API calls is slow. They're a contract. `input_image` beats `image2`. Note the `deployment_id`. That's what your code stores. Confirm one run end to end before wiring it into anything. For breaking changes, deploy alongside and migrate — don't mutate what's live. # Pix Agent Source: https://support.myapps.ai/create/agent Create by describing what you want, in conversation **Pix Agent** is Pixio as a chat. Describe what you want and it picks the model, runs the generation, and shows the result inline — no forms, no model list. ## When to use it | Use Pix Agent when | Use [Generate](/generate) when | | ------------------------------------- | --------------------------------------- | | You don't know which model fits | You know the model you want | | You want to iterate in plain language | You need exact control of every input | | You're exploring an idea | You're producing a specific deliverable | ## How it works Ask for what you want — an image, a clip, a voice track. You can attach an image as a reference. Pix Agent chooses a suitable model and runs it. Long renders show as still rendering and appear in the conversation when they finish. Follow up conversationally — "same scene, new angle", "same vibe, slower tempo", "try a cinematic style". Pix Agent carries the context forward. ## Attaching images Attach an image to use it as a reference — a starting frame, a style guide, a character. Uploads join your library like any other file. ## Conversations Conversations are saved, so you can leave and come back to one. Each keeps its own context and generated media. ## When something fails A failed generation is called out in the conversation and can be retried — sometimes with a more reliable model. **Failed runs aren't charged.** ## Where results go Everything Pix Agent makes is a normal Pixio generation: * Appears in **[My Assets](/assets)** * Prompts saved to **[Prompt History](/library/prompt-history)** * Costs the same credits as it would in Generate * Can be downloaded, published, or chained onward in Generate Explore in Pix Agent, then finish in Generate. Once you know which model and settings you want, Generate gives you exact control over inputs and seeds. Model availability follows your plan here too — Pix Agent won’t pick a model your plan can't run. # Boards Source: https://support.myapps.ai/create/boards Node workspaces for multi-step media pipelines **Boards** is a node workspace. You place nodes on a canvas, connect them into branches, and route shots, references, and multi-step media pipelines — the flow map is the source of truth. ## Boards vs Canvas Both are visual workspaces. They solve different problems. | | Boards | [Canvas](/create/canvas) | | ----------- | -------------------------------- | -------------------------- | | Model | Nodes and connections | Layered document | | Thinking in | Flow and branches | Composition and stacking | | Best for | Multi-step pipelines, variations | Compositing a single image | | Node graphs | Yes — that's the point | **No** | If you're chaining steps, use Boards. If you're building one image out of layers, use Canvas. ## Working in it From the library, start a new board or open an existing one. Each board is its own saved node workspace. Place nodes, connect them into branches, and route shots, references, and multi-step pipelines. Use the board hub and director tools to run generations, preview outputs, and keep the flow map authoritative. ## What's on a board | Piece | What it's for | | ------------------ | ------------------------------------------------------------ | | **Nodes** | Steps in the pipeline — a generation, a reference, an output | | **Connections** | How output flows into the next step | | **Branches** | Explore variations side by side without losing the original | | **Asset hub** | Bring existing assets into the board | | **Director panel** | Direct the shot rather than hand-tuning raw parameters | | **Assembly strip** | Line up outputs toward a finished sequence | ## Why branch Branching is the reason to use a board. Instead of overwriting a prompt and losing what you had, you fork — three treatments of the same shot side by side, all visible at once, all traceable back to the same source. Branch at the point of uncertainty. If you're unsure about a style, fork there and generate both — you'll see the answer instead of arguing with yourself about it. ## Sharing Boards can be shared by link, so someone can see the flow and its outputs without a Pixio account. See [Share links](/library/share-links). ## Credits Nodes that generate cost credits at their model's normal rate. Placing and connecting nodes is free — you only pay when a generation runs, so map the whole flow before running any of it. Board and spatial-canvas project CRUD is available through the API. See [what you can automate](/integrations/what-you-can-automate). # Canvas Source: https://support.myapps.ai/create/canvas A layered compositing workspace — no node graphs **Canvas** is a compositing-style workspace. You stack images, frames, and adjustments in layers, position them precisely, preview, and export — built for visual iteration, without node graphs. This is Canvas in the Pixio web app. It's different from **[Pixio Canvas for iPad](/pixio-canvas/support)**, the drawing and tracing companion app. ## Canvas vs Boards | | Canvas | [Boards](/create/boards) | | ----------- | ------------------------ | ------------------------ | | Model | Layered document | Nodes and connections | | Thinking in | Composition and stacking | Flow and branches | | Best for | Building one image | Multi-step pipelines | | Node graphs | **No** | Yes | Reach for Canvas when the job is *make this one image right*. ## Working in it From the library, create a new canvas or resume one. Each canvas stores its own layered document and edits. Stack images, frames, and adjustments in a familiar compositing-style workspace. Position elements precisely, preview the composition, and export when you're happy. ## What you work with | Piece | What it does | | ------------------ | --------------------------------------------------- | | **Layers** | Stacked elements — images, frames, adjustments | | **Frames** | Define the composition area | | **Adjustments** | Non-destructive changes over what's beneath | | **Templates** | Start from a prepared setup instead of blank | | **Director panel** | Describe the change you want instead of hand-tuning | | **Frames preview** | See the composition as it will export | ## Projects are saved Each canvas keeps its own document and edit history, so you can leave a composition half-finished and pick it up later. Everything lives in your canvas library. ## Sharing A canvas can be shared by link for review without a Pixio account. See [Share links](/library/share-links). ## Credits Composing, positioning, and exporting are free. Generating new imagery from the canvas costs credits at the model's normal rate. Generate elements at the size you'll actually use. Compositing a small image into a large frame and scaling it up loses detail that no amount of layer work brings back. Canvas design CRUD and operations are available through the API. See [what you can automate](/integrations/what-you-can-automate). # Cinema Studio Source: https://support.myapps.ai/create/cinema-studio Plan shots, direct the camera, and generate in place **Cinema Studio** is for planning a video before you pay to generate it. You lay out shots, direct the camera, and generate frames in place — the storyboard stays the source of truth. ## What it gives you Turns a plain description into proper shot language — framing, lens, movement. Each frame carries its own references, so look is controlled per shot instead of globally. Generate a frame directly on the board and keep working — the result returns to the board. Use your phone as a physical camera with a live viewfinder. See [CamView](/advanced/camview). Keyframe camera moves and figure animation rather than hoping a prompt implies motion. Describe the look as a director; it becomes a cinematographer's prompt. ## Why plan first Video is the most expensive output in Pixio, and prompt-only direction is the least predictable way to get a specific shot. Deciding framing and camera movement *before* generating means you buy the shot you meant instead of iterating at video prices. Storyboard the sequence, generate stills to confirm composition, then commit to video. Two cheap steps to avoid one expensive mistake. ## Storyboards Storyboards are saved separately and managed on their own page — see [Storyboards](/create/storyboards). ## Where it fits | Surface | Role | | -------------------------------------- | ---------------------------------------------- | | **Cinema Studio** | Plan and direct the shots | | **[Video Agent](/create/video-agent)** | Generate an episode with character consistency | | **[Video Editor](/video-editor)** | Timeline assembly, captions, export | | **[Generate](/generate)** | One clip, full manual control | Storyboard CRUD and prompt-to-storyboard direction are available through the API. See [what you can automate](/integrations/what-you-can-automate). # Storyboards Source: https://support.myapps.ai/create/storyboards Plan a sequence frame by frame before generating it Storyboards are the planning layer of [Cinema Studio](/create/cinema-studio). A storyboard is a saved sequence of frames — each with its own description, shot language, and reference images — that you build before spending credits on video. ## Why storyboard Video is the most expensive output in Pixio, and the least predictable from a prompt alone. A storyboard front-loads the decisions that are cheap to change: * What's in each shot * How it's framed * What it references * What order it goes in Change a frame's framing in a storyboard: free. Discover the framing was wrong after generating six clips: expensive. ## What a frame holds | | | | -------------------- | -------------------------------------------------------------------------------------- | | **Description** | What happens in the shot | | **Shot language** | Framing, lens, movement — generated from your description if you'd rather not write it | | **Reference images** | Per frame, so look is controlled shot by shot rather than globally | Per-frame references are the feature to exploit. A single global style reference flattens a sequence; per-frame references let a scene shift look deliberately while staying coherent. ## Working with them Lay out frames in order. Describe each shot plainly — shot language can be generated from the description. Add reference images per frame for subject and look. Generate a frame from the board and the result comes back to it, so the storyboard stays the source of truth. Once the frames are right, generate the clips — via [Video Agent](/create/video-agent) for episodic work with character memory, or [Generate](/generate) for single shots. ## Where storyboards live Storyboards are saved and listed on their own page, separate from timeline projects. Storyboards intentionally don't appear in the [Video Editor](/video-editor) library — they're plans, not footage. The editor works with clips. ## Related * [Cinema Studio](/create/cinema-studio) — the studio around storyboards * [CamView](/advanced/camview) — block a specific shot in 3D * [Video Agent](/create/video-agent) — generate a multi-segment episode * [what you can automate](/integrations/what-you-can-automate) — storyboard CRUD and prompt-to-storyboard direction are scriptable # Video Agent Source: https://support.myapps.ai/create/video-agent Reference-driven episodic video with character memory and stitched exports **Video Agent** builds multi-segment videos. You give it a brief and references; it generates the segments, keeps your characters consistent across them, and stitches the result into one episode. ## What makes it different | | Generate | Video Agent | | ----------- | ---------------- | ------------------------------------ | | Output | One clip per run | A multi-segment episode | | Consistency | You manage it | **Character memory** across segments | | Assembly | You do it | Stitched export | | Best for | A single shot | A sequence that has to hang together | ## Core ideas Characters persist across segments so a face or costume doesn't drift shot to shot — the hardest problem in multi-shot AI video. References steer look and subject rather than relying on prompt wording alone. Each segment generates independently, so you can regenerate one without redoing the episode. Approved segments assemble into a single video. ## Working in it Projects are saved and resumable — episodic work spans sessions. Set these before generating in bulk. Consistency comes from the references, so getting them right first saves the most credits. Review each one. Regenerate the weak ones individually rather than restarting. Stitch the approved segments into the finished video. ## Credits Every segment is a video generation and costs accordingly — an episode is the sum of its segments. Regenerating one segment costs one segment. Video is the most expensive thing in Pixio. Get your references and first segment right before generating the rest, or you'll pay for a full episode twice. ## Where it fits * **[Cinema Studio](/create/cinema-studio)** — plan shots as a storyboard before generating * **[Video Editor](/video-editor)** — a real timeline for precise trimming, captions, and layered edits * **[Generate](/generate)** — single clips with full control Autonomous orchestration and final assembly aren't available through the API — Video Agent projects can be persisted programmatically, but the end-to-end run happens in the app. See [what you can automate](/integrations/what-you-can-automate). # Credits & Plans Source: https://support.myapps.ai/credits What generations cost, what each plan includes, and how billing works Pixio bills with **credits**. Every generation costs credits; your **plan** decides how many you get each month and which models you can reach. ## How credits work Each model has its own price. It's shown in the model list and again before you submit, so you always know the cost up front. If a generation fails, you aren't charged. You only pay for output you actually get. What drives the cost: * **Output type** — images are cheap, video is expensive, 3D and audio sit in between * **Duration** — video and audio models are often billed per second * **Resolution and quality** — higher settings cost more on most models * **Model tier** — a fast draft model and a flagship model differ by a lot Iterate on cheap image models, then chain the result you like into an expensive video model. See [chained actions](/generate#chained-actions) — it's the single biggest credit saver in Pixio. ## Free credits Every new account gets **50 credits that never expire**, with access to all major models. No card required to start. ## Plans | Plan | Monthly | Yearly | Credits | Notable | | -------------- | -------- | ---------- | ---------------- | ------------------------------------------- | | **Free** | \$0 | \$0 | 50, never expire | Access to all major models | | **Starter** | \$15.99 | \$159.99 | 2,000 / month | All models · bespoke prompt optimizer | | **Pro** | \$29.99 | \$299.99 | 3,500 / month | Everything in Starter | | **Premium** | \$59.99 | \$599.99 | 7,500 / month | **LoRA training** · beta features | | **Maker Mode** | \$199.99 | \$1,999.99 | 15,000 | **Unlimited generations** · unlimited LoRAs | Yearly billing works out to roughly ten months for twelve. ### Maker Mode Maker Mode is the "stop counting" tier — **unlimited generations** on a wide set of models plus **unlimited LoRAs**. Instead of drawing down a balance on every run, eligible models are covered by the plan. Unlimited applies to the models included in Maker Mode, not literally every model in the catalog. The app shows which models are covered. ### LoRA training Training your own custom models requires **Premium** or above. Maker Mode removes the cap on how many you can have. See Model Training in the app. ## Credit packs One-time packs add **permanent credits** that don't expire with your billing cycle: | Pack | Price | | -------------- | ----- | | 400 credits | \$5 | | 1,000 credits | \$10 | | 2,300 credits | \$20 | | 5,000 credits | \$40 | | 12,500 credits | \$100 | Useful for a busy month without changing your subscription. Packs stack on top of your monthly allowance. ## Running out When your balance won't cover a run, Pixio tells you **before** starting it — nothing gets queued and nothing is charged. From there you can upgrade your plan or buy a credit pack on the **Billing** page. ## Billing page Open **Billing** from your profile menu to see your plan, your current balance, and your payment details, and to change plans or cancel. ## Personal vs team | | Credits | Assets | | -------------------- | ---------------------------------- | ------------------------ | | **Personal account** | Your own balance | Private to you | | **Team account** | One shared pool for the whole team | Shared with every member | A team account bills separately from your personal account. Team billing is usually managed by an admin — see [Teams & Account](/billing/account). Prices and credit allowances here reflect the current plans. The Billing page in the app is always authoritative for your account. # Pixio API Development Source: https://support.myapps.ai/development Learn how to interact with the Pixio API to trigger workflows and upload files **Prerequisite**: You should have Node.js (version 18.10.0 or higher) and an API token from the Pixio dashboard. ### Step 1: Install the Pixio API SDK To start interacting with the Pixio API, you need to install the Pixio SDK using npm or yarn: ```bash npm theme={null} npm install @pixio/api-sdk ` `yarn add @pixio/api-sdk` ``` ### Step 2: Configure API Key Once the SDK is installed, create a `.env` file in your project to securely store your Pixio API key. ```bash npm theme={null} `PIXIO_API_KEY=your-api-key-here` ``` ### Step 3: Set up the Pixio API Client Next, set up the Pixio API client in your code: ```bash npm theme={null} `const PixioAPI = require('@pixio/api-sdk'); const client = new PixioAPI(process.env.PIXIO_API_KEY); // Example: Fetch workflow projects client.getProjects().then(projects => { console.log(projects); });` ``` ### Step 4: Running a Workflow You can trigger a workflow by making a POST request to the `/run` endpoint with the deployment ID and input data: ```bash npm theme={null} `client.runWorkflow({ deployment_id: "your-deployment-id", inputs: { input_text: "Example input", input_image: "https://example.com/image.png" } }).then(response => { console.log('Run ID:', response.run_id); });` ``` ### Step 5: Retrieve Workflow Output After triggering the workflow, you can retrieve the output using the `run_id`: ````bash npm theme={null} `client.getWorkflowOutput("your-run-id").then(output => { console.log('Output:', output); });` ### Step 6: Uploading Files To upload a file (image, document, etc.), you need to first get an upload URL: ```bash npm `client.getUploadUrl({ type: "image/png", file_size: 102400 }).then(response => { console.log('Upload URL:', response.upload_url); console.log('File ID:', response.file_id); });` ```` ### Custom Ports Pixio API calls are generally executed on standard ports. If you are working in a local development environment and need to customize the port for running your Node.js app: ```bash npm theme={null} `const PORT = process.env.PORT || 3333; app.listen(PORT, () => console.log(`Server running on port ${PORT}`));` ``` ### Common Errors If you encounter any errors such as `EADDRINUSE` or other port-related issues, make sure that no other services are running on the same port. ```bash npm theme={null} `Error: listen EADDRINUSE: address already in use :::3000` ``` ## Pixio API Versions To ensure compatibility with the latest Pixio API features, keep the SDK up-to-date: ```bash npm theme={null} `npm update @pixio/api-sdk` `yarn upgrade @pixio/api-sdk` ``` ## Deployment Once you've built your Pixio-powered application, you can deploy it on platforms like Vercel, Heroku, or AWS. For more details, check the [Pixio API Documentation](https://support.myapps.ai/api-reference/endpoint/get). If you have any issues please contact us. ## Troubleshooting If you run into any issues while working with the Pixio API, here are a few troubleshooting steps: Ensure your `.env` file is correctly set up with your Pixio API key. Regenerate the API key if needed from the dashboard. Make sure your token is valid and hasn't expired. You can renew it in the Pixio API dashboard under API Keys. If you receive unexpected responses from the API, try resetting your environment and reinstalling dependencies by running `npm install` or `yarn install`. For further details on the Pixio API, check out the API Reference. # 3D & utility models Source: https://support.myapps.ai/families/3d-and-utility Smaller families covering 3D helpers and single-purpose media utilities — background removal, cleanup, and video operations.. Smaller families covering 3D helpers and single-purpose media utilities — background removal, cleanup, and video operations. ## Which one to pick These do one job each. Pick by the operation you need rather than by family. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** video-to-video (5) · image-to-3D (4) · text-to-3D (2) · image editing (2) **Price range:** 5–75 credits per run **13 models in Pixio.** Credits are per run unless a unit is shown. | Family | Model | Does | Credits | Access | | ----------- | ------------------------------ | ------------- | ------- | ------------- | | 3D Tools | TripoSplat | Image → 3D | 5 | All plans | | 3D Tools | Rodin V2.5 Fast (Image to 3D) | Image → 3D | 13 | All plans | | 3D Tools | Trellis 2 Retexture | Image → 3D | 40 | All plans | | 3D Tools | Hyper3D Rodin V2.5 Image to 3D | Image → 3D | 50 | All plans | | ClarityAI | Crystal Image Upscaler | Image → image | 20 | All plans | | RemoveBG | Remove Background | Image → image | 75 | All plans | | 3D Tools | Rodin V2.5 Fast (Text to 3D) | Text → 3D | 13 | All plans | | 3D Tools | Hyper3D Rodin V2.5 Text to 3D | Text → 3D | 50 | All plans | | Video Tools | Caption Video | Video → video | 10 | Free on Maker | | ClarityAI | Crystal Video Upscaler | Video → video | 10 | All plans | | Video Tools | Upscale Video | Video → video | 10 | All plans | | Video Tools | Video Background Remover | Video → video | 10 | Free on Maker | | Pixelcut | Video Background Removal | Video → video | 32 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # More audio & avatar models Source: https://support.myapps.ai/families/audio-avatar-roundup Smaller audio, music, and avatar families beyond the main options.. Smaller audio, music, and avatar families beyond the main options. ## Which one to pick Worth comparing on the same input — audio quality is very subjective. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** audio generation (5) · audio transformation (4) · video-to-video (1) · image editing (1) · text-to-video (1) · audio-driven video (1) · lipsync (1) **Price range:** 1–100 credits per run **14 models in Pixio.** Credits are per run unless a unit is shown. | Family | Model | Does | Credits | Access | | ------- | ----------------------------- | ------------- | ---------------- | --------- | | Cohere | Cohere Transcribe | Audio → audio | 1 | All plans | | xAI | xAI Speech to Text | Audio → audio | 1 | All plans | | Mirelo | Mirelo SFX 1.6 Inpaint Audio | Audio → audio | 7 | All plans | | Mirelo | Mirelo SFX 1.6 Extend Audio | Audio → audio | 16 | All plans | | Argil | Argil Avatars Audio-to-Video | Audio → video | 6 | All plans | | Argil | Argil Avatars Train | Image → image | 100 | All plans | | DaVinci | DaVinci MagiHuman | Lipsync | 25 | All plans | | Mureka | Mureka Create (AI Lyrics) | Text → audio | 15 | All plans | | Mureka | Mureka Create Instrumental | Text → audio | 15 | All plans | | Mirelo | Mirelo SFX 1.6 Text to Audio | Text → audio | 16 | All plans | | Mureka | Mureka Create (Advanced) | Text → audio | 20 | All plans | | Sonilo | Sonilo V1.1 Text to Music | Text → audio | 22 / 0.24 second | All plans | | Argil | Argil Avatars Text-to-Video | Text → video | 6 | All plans | | Mirelo | Mirelo SFX 1.6 Video to Video | Video → video | 16 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Bria Source: https://support.myapps.ai/families/bria Image family with text-to-image, editing, and a video model, aimed at commercial use.. Image family with text-to-image, editing, and a video model, aimed at commercial use. ## Which one to pick Worth a look for product and commercial imagery. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (3) · image editing (3) · video-to-video (1) **Price range:** 2–61 credits per run **7 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | --------------------- | ------------- | ------- | ------------- | | Bria Extract Object | Image → image | 3 | All plans | | Bria GenFill | Image → image | 6 | All plans | | Bria Upscale Creative | Image → image | 6 | All plans | | Fast | Text → image | 2 | Free on Maker | | Base | Text → image | 4 | Free on Maker | | HD | Text → image | 4 | Free on Maker | | Bria Video BG Removal | Video → video | 61 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # ByteDance (Seedance & Seedream) Source: https://support.myapps.ai/families/bytedance Video-led family with strong image models alongside it — text-to-video and image-to-video plus text-to-image, editing, video-to-video, and lipsync.. Video-led family with strong image models alongside it — text-to-video and image-to-video plus text-to-image, editing, video-to-video, and lipsync. ## Which one to pick Video models here vary a lot in cost by duration and resolution. Test at the shortest length the model allows before scaling up. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-video (18) · image-to-video (8) · text-to-image (6) · image editing (4) · video-to-video (2) · audio generation (2) · lipsync (1) **Price range:** 3–390 credits per run **41 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ------------------------------------- | ------------- | ------- | ------------- | | Seedream v4 Edit | Image → image | 3 | Free on Maker | | Bernini-R Edit Image | Image → image | 5 | All plans | | Seedream 5 Edit | Image → image | 5 | Free on Maker | | Seedream v4.5 Edit | Image → image | 5 | Free on Maker | | Seedance v1 Lite Image to Video | Image → video | 5 | Free on Maker | | Seedance v1 Lite Reference to Video | Image → video | 5 | Free on Maker | | Seedance v1 Pro Fast Image to Video | Image → video | 10 | Free on Maker | | Seedance v1 Pro Image to Video | Image → video | 10 | Free on Maker | | Bernini-R Reference to Video | Image → video | 39 | All plans | | Seedance 2 Mini Image to Video | Image → video | 100 | All plans | | Seedance 2 Fast (Face) Image to Video | Image → video | 210 | All plans | | Seedance 2 (Face) Image to Video | Image → video | 240 | All plans | | OmniHuman v1.5 | Lipsync | 10 | Free on Maker | | Seed Speech TTS v2 | Text → audio | 3 | All plans | | Seed Audio 1.0 | Text → audio | 25 | All plans | | Dreamina v3.1 | Text → image | 3 | Free on Maker | | Seedream v4 | Text → image | 3 | Free on Maker | | Seedream 5 | Text → image | 5 | Free on Maker | | Seedream 5 Maker | Text → image | 5 | Free on Maker | | Seedream v4.5 | Text → image | 5 | Free on Maker | | Seedream 5 Pro | Text → image | 10 | Free on Maker | | Seedance v1 Lite Text to Video | Text → video | 5 | Free on Maker | | Seedance v1 Pro Fast Text to Video | Text → video | 10 | Free on Maker | | Seedance v1 Pro Text to Video | Text → video | 10 | Free on Maker | | Bernini-R Text to Video | Text → video | 39 | All plans | | Seedance 2 Maker | Text → video | 40 | Free on Maker | | Seedance 2 Mini Stable | Text → video | 64 | Maker | | Seedance 2 Mini Omni | Text → video | 100 | All plans | | Seedance 2 Mini Text to Video | Text → video | 100 | All plans | | Seedance 2 Fast Omni Stable | Text → video | 103 | All plans | | Seedance 2 Max | Text → video | 125 | All plans | | Seedance 2 Omni Stable | Text → video | 128 | All plans | | Seedance 2 Direct | Text → video | 156 | All plans | | Seedance 2 Fast Maker | Text → video | 175 | Free on Maker | | Seedance 2 Fast (Face) Omni | Text → video | 210 | All plans | | Seedance 2 Fast (Face) Text to Video | Text → video | 210 | All plans | | Seedance 2 (Face) Omni | Text → video | 240 | All plans | | Seedance 2 (Face) Text to Video | Text → video | 240 | All plans | | Seedance 2.5 Preview (Early Release) | Text → video | 390 | All plans | | Bernini-R Edit Video | Video → video | 39 | All plans | | Bernini-R Reference Edit Video | Video → video | 39 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # ElevenLabs Source: https://support.myapps.ai/families/elevenlabs Voice and audio — text-to-speech and audio transformation, including voice work.. Voice and audio — text-to-speech and audio transformation, including voice work. ## Which one to pick The default for speech and narration. For music, see SongCraft or MiniMax. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** audio generation (4) · audio transformation (2) **Price range:** 0–24 credits per run **6 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ----------------- | ------------- | ------- | ------------ | | Voice Clone (IVC) | Audio → audio | 0 | Free on Free | | Scribe v2 | Audio → audio | 1 | All plans | | Text to Speech | Text → audio | 8 | All plans | | Sound Effects | Text → audio | 10 | Free on Free | | Text to Dialogue | Text → audio | 10 | All plans | | Music (Compose) | Text → audio | 24 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Flux Source: https://support.myapps.ai/families/flux The workhorse image family: text-to-image, image editing, and inpainting. The workhorse image family: text-to-image, image editing, and inpainting. Strong prompt adherence and the widest LoRA support in Pixio, which makes it the family to train custom models against. ## Which one to pick Draft with a fast variant, finish with a higher-quality one. If you have trained a LoRA, check which Flux models accept custom weights before committing to a look. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (17) · image editing (13) · inpainting (3) **Price range:** 1–15 credits per run **33 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ------------------------- | ------------- | ------- | ------------------------------------ | | Flux 2 Flash Editing | Image → image | 2 | Free on Maker | | Flux 2 Klein 4B Edit | Image → image | 3 | Free on Maker | | Flux 2 Turbo Editing | Image → image | 4 | Free on Maker | | Kontext Pro Editing | Image → image | 4 | Free on Maker | | Kontext Pro Editing Multi | Image → image | 4 | Free on Maker | | Flux 2 Klein 9B Edit | Image → image | 5 | Free on Maker | | Kontext Max Editing | Image → image | 8 | Free on Maker | | Kontext Max Editing Multi | Image → image | 8 | Free on Maker | | Flux 2 Edit | Image → image | 10 | Free on Maker | | Flux 2 Flex Edit | Image → image | 10 | Free on Maker | | Flux 2 Pro Edit | Image → image | 10 | Free on Maker | | Flux 2 Max Edit | Image → image | 15 | Free on Maker | | Flux 2 Pro Outpaint | Image → image | 15 | All plans | | Flux Dev Inpainting | Inpainting | 3 | Free on Maker, Premium, Pro, Starter | | Flux Pro Fill | Inpainting | 5 | Free on Maker | | Flux Pro Fill Finetuned | Inpainting | 7 | Free on Maker | | Flux Schnell | Text → image | 1 | Free on Maker, Premium, Pro, Starter | | Flux 2 Flash | Text → image | 2 | Free on Maker | | Flux 2 Klein 4B | Text → image | 3 | Free on Maker | | Flux Dev | Text → image | 3 | Free on Maker, Premium, Pro, Starter | | Flux Krea | Text → image | 3 | Free on Maker | | Flux SRPO | Text → image | 3 | Free on Maker | | Flux 2 Turbo | Text → image | 4 | Free on Maker | | Kontext Pro | Text → image | 4 | Free on Maker | | Flux 2 Klein 9B | Text → image | 5 | Free on Maker | | Flux Pro | Text → image | 5 | Free on Maker | | Flux Pro Ultra | Text → image | 7 | Free on Maker | | Flux Pro Ultra Finetuned | Text → image | 7 | Free on Maker | | Kontext Max | Text → image | 8 | Free on Maker | | Flux 2 | Text → image | 10 | Free on Maker | | Flux 2 Flex | Text → image | 10 | Free on Maker | | Flux 2 Pro | Text → image | 10 | Free on Maker | | Flux 2 Max | Text → image | 15 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Google (Veo & Imagen) Source: https://support.myapps.ai/families/google Covers all four output types — video, image, image editing, and audio. Covers all four output types — video, image, image editing, and audio. Veo handles video, Imagen handles stills. ## Which one to pick A good default when you want one family across a whole project. Check per-model pricing carefully; the range within this family is wide. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (8) · text-to-image (7) · text-to-video (4) · image editing (3) · audio generation (3) · video-to-video (2) **Price range:** 2–400 credits per run **27 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ----------------------------------------- | ------------- | ------- | ------------------------------------ | | Nano-Banana Edit | Image → image | 4 | Free on Maker | | Nano-Banana 2 Edit | Image → image | 7 | Free on Maker | | Nano-Banana Pro Edit | Image → image | 15 | All plans | | Veo 3.1 Lite Preview (First & Last Frame) | Image → video | 32 | Free on Maker | | Veo 3.1 Lite Preview (Image to Video) | Image → video | 32 | Free on Maker | | Veo 3.1 Fast First–Last Frame to Video | Image → video | 200 | All plans | | Veo 3.1 Fast Image to Video | Image → video | 200 | All plans | | Veo 3.1 Fast Reference to Video | Image → video | 200 | All plans | | Veo 3.1 First–Last Frame to Video | Image → video | 400 | All plans | | Veo 3.1 Image to Video | Image → video | 400 | All plans | | Veo 3.1 Reference to Video | Image → video | 400 | All plans | | Lyria 3 Clip | Text → audio | 5 | Free on Maker | | Lyria 3 Pro | Text → audio | 15 | Free on Maker | | Gemini 3.1 Flash TTS Preview | Text → audio | 30 | All plans | | Imagen 4 Fast | Text → image | 2 | Free on Maker, Premium, Pro, Starter | | Imagen 4 | Text → image | 4 | Free on Maker | | Nano-Banana | Text → image | 4 | Free on Maker | | Nano-Banana Lite | Text → image | 4 | Free on Maker | | Imagen 4 Ultra | Text → image | 6 | Free on Maker | | Nano-Banana 2 | Text → image | 7 | Free on Maker | | Nano-Banana Pro | Text → image | 15 | All plans | | Veo 3.1 Lite Preview (Text to Video) | Text → video | 32 | Free on Maker | | Gemini Omni Flash | Text → video | 100 | All plans | | Veo 3.1 Fast Text to Video | Text → video | 200 | All plans | | Veo 3.1 Text to Video | Text → video | 400 | All plans | | Veo 3.1 Fast Extend Video | Video → video | 105 | All plans | | Veo 3.1 Extend Video | Video → video | 280 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Grok Source: https://support.myapps.ai/families/grok Compact family spanning image, image editing, video, and video editing — including the instruction-driven Edit Image and Edit Video actions surfaced in Generate.. Compact family spanning image, image editing, video, and video editing — including the instruction-driven Edit Image and Edit Video actions surfaced in Generate. ## Which one to pick Best for describing a change to existing media rather than generating from scratch. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (4) · text-to-image (2) · image editing (2) · text-to-video (2) · video-to-video (2) **Price range:** 5–160 credits per run **12 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ----------------------------------- | ------------- | ------- | ------------- | | Grok Imagine Image Edit | Image → image | 5 | Free on Maker | | Grok Imagine Quality Edit | Image → image | 7 | Free on Maker | | Grok Imagine Image-to-Video | Image → video | 32 | Free on Maker | | Grok Imagine Reference to Video | Image → video | 80 | All plans | | Grok Imagine 1.5 Image-to-Video | Image → video | 133 | All plans | | Grok Imagine 1.5 Reference-to-Video | Image → video | 133 | All plans | | Grok Imagine Text-to-Image | Text → image | 5 | Free on Maker | | Grok Imagine Quality | Text → image | 7 | Free on Maker | | Grok Imagine Text-to-Video | Text → video | 60 | Free on Maker | | Grok Imagine 1.5 Text-to-Video | Text → video | 133 | All plans | | Grok Imagine Extend Video | Video → video | 60 | All plans | | Grok Imagine Video Edit | Video → video | 160 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # HeyGen Source: https://support.myapps.ai/families/heygen Avatar and presenter video — text-to-video, video-to-video, and image-to-video aimed at talking-head output.. Avatar and presenter video — text-to-video, video-to-video, and image-to-video aimed at talking-head output. ## Which one to pick Use for a presenter or spokesperson. For syncing an existing face to audio, compare with Sync. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-video (4) · video-to-video (2) · image-to-video (1) **Price range:** 5–22 credits per run **7 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ------------------------------ | ------------- | ------- | --------- | | HeyGen Avatar 4 Image-to-Video | Image → video | 10 | All plans | | HeyGen Avatar 3 Digital Twin | Text → video | 5 | All plans | | HeyGen Avatar 4 Digital Twin | Text → video | 10 | All plans | | HeyGen Video Agent | Text → video | 10 | All plans | | HeyGen Avatar 5 Digital Twin | Text → video | 22 | All plans | | HeyGen Translate Speed | Video → video | 5 | All plans | | HeyGen Translate Precision | Video → video | 10 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Hunyuan Source: https://support.myapps.ai/families/hunyuan The broadest 3D family in Pixio — text-to-3D, image-to-3D, and 3D refinement — with image, video, and lipsync models alongside.. The broadest 3D family in Pixio — text-to-3D, image-to-3D, and 3D refinement — with image, video, and lipsync models alongside. ## Which one to pick For 3D, decide whether you are starting from a prompt or an image; that picks the model. Refinement models improve a mesh you already have. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-3D (5) · image-to-3D (4) · 3D refinement (2) · image-to-video (2) · text-to-image (1) · text-to-video (1) · video-to-video (1) · lipsync (1) **Price range:** 10–280 credits per run **17 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | --------------------------------- | ------------- | ------- | ------------- | | Hunyuan 3D V3.1 Segment Model | 3D → 3D | 50 | Free on Maker | | Hunyuan 3D V3.1 Optimize Model | 3D → 3D | 100 | All plans | | Hunyuan 3D V3.1 Rapid Image to 3D | Image → 3D | 50 | Free on Maker | | Hunyuan 3D V3 Image to 3D | Image → 3D | 100 | All plans | | Hunyuan 3D V3 Sketch to 3D | Image → 3D | 100 | All plans | | Hunyuan 3D V3.1 Pro Image to 3D | Image → 3D | 100 | All plans | | Hunyuan Video Image to Video | Image → video | 40 | Free on Maker | | Hunyuan Custom | Image → video | 240 | All plans | | Hunyuan Avatar | Lipsync | 280 | All plans | | Hunyuan Motion Fast | Text → 3D | 10 | Free on Maker | | Hunyuan Motion | Text → 3D | 15 | Free on Maker | | Hunyuan 3D V3.1 Rapid Text to 3D | Text → 3D | 50 | Free on Maker | | Hunyuan 3D V3 Text to 3D | Text → 3D | 100 | All plans | | Hunyuan 3D V3.1 Pro Text to 3D | Text → 3D | 100 | All plans | | Hunyuan Image V3 | Text → image | 10 | Free on Maker | | Hunyuan Video Text to Video | Text → video | 40 | Free on Maker | | Hunyuan Video to Video | Video → video | 40 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Ideogram Source: https://support.myapps.ai/families/ideogram Image family weighted toward editing, with text-to-image alongside. Image family weighted toward editing, with text-to-image alongside. Best known for handling text inside images. ## Which one to pick The family to try when your image needs legible words in it. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image editing (7) · text-to-image (5) **Price range:** 2–7 credits per run **12 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------------------- | ------------- | ------- | ------------- | | Ideogram V4 Image to Image | Image → image | 2 | All plans | | Remove Background | Image → image | 2 | Free on Maker | | Remix (V4) | Image → image | 5 | Free on Maker | | Edit (V3) | Image → image | 6 | Free on Maker | | Reframe (V3) | Image → image | 6 | Free on Maker | | Remix (V3) | Image → image | 6 | Free on Maker | | Replace Background (V3) | Image → image | 6 | Free on Maker | | Generate (V4) | Text → image | 5 | Free on Maker | | Ideogram V4 Instant | Text → image | 5 | Free on Maker | | Generate (V3) | Text → image | 6 | Free on Maker | | Ideogram V4 Tiling | Text → image | 6 | All plans | | Ideogram V4 Fast | Text → image | 7 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # More image models Source: https://support.myapps.ai/families/image-roundup Smaller image families worth knowing about, including virtual try-on and a few specialist generators.. Smaller image families worth knowing about, including virtual try-on and a few specialist generators. ## Which one to pick Try these when the larger families are not giving you the look you want. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (14) · image editing (10) **Price range:** 2–40 credits per run **24 models in Pixio.** Credits are per run unless a unit is shown. | Family | Model | Does | Credits | Access | | -------------- | --------------------------- | ------------- | ------- | ------------- | | ImagineArt | ImagineArt 2.0 Edit Preview | Image → image | 4 | All plans | | Virtual Try On | Virtual Try On | Image → image | 4 | Free on Maker | | Microsoft | MAI Image 2.5 Edit | Image → image | 5 | All plans | | Virtual Try On | Fashion Photoshoot | Image → image | 7 | Free on Maker | | Virtual Try On | Fashn Tryon v1.6 | Image → image | 8 | Free on Maker | | HiDream | HiDream O1 Image Dev Edit | Image → image | 20 | All plans | | HiDream | HiDream O1 Image Edit | Image → image | 20 | All plans | | Phota | Phota Edit | Image → image | 20 | All plans | | Phota | Phota Enhance | Image → image | 25 | All plans | | Microsoft | MAI Image 2.5 Pro Edit | Image → image | 35 | All plans | | Krea | Krea 2 Medium Turbo | Text → image | 2 | All plans | | Krea | Krea 2 Turbo | Text → image | 2 | All plans | | Krea | Krea 2 Turbo Style | Text → image | 2 | All plans | | Sourceful | Riverflow 2.0 Fast | Text → image | 5 | Free on Maker | | Krea | Krea 2 Medium | Text → image | 6 | All plans | | Microsoft | MAI Image 2.5 | Text → image | 6 | Free on Maker | | Krea | Krea 2 Large | Text → image | 11 | All plans | | HiDream | HiDream O1 Image | Text → image | 20 | All plans | | HiDream | HiDream O1 Image Dev | Text → image | 20 | All plans | | Phota | Phota | Text → image | 20 | All plans | | Sourceful | Riverflow 2.0 Pro | Text → image | 20 | All plans | | Sourceful | Riverflow 2.5 Fast | Text → image | 20 | All plans | | Microsoft | MAI Image 2.5 Pro | Text → image | 25 | All plans | | Sourceful | Riverflow 2.5 Pro | Text → image | 40 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Image tools Source: https://support.myapps.ai/families/image-tools Single-purpose image utilities — the operations behind several chained actions in Generate.. Single-purpose image utilities — the operations behind several chained actions in Generate. ## Which one to pick Not for generating from scratch. Use when you need one specific operation applied. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image editing (5) **Price range:** 0–5 credits per run **5 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ------------------ | ------------- | ------- | ------------------------------------ | | Background Removal | Image → image | 0 | All plans | | Photo Restoration | Image → image | 4 | Free on Maker | | Advanced Face Swap | Image → image | 5 | All plans | | Face Swap | Image → image | 5 | All plans | | Upscale | Image → image | 5 | Free on Maker, Premium, Pro, Starter | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Kling Source: https://support.myapps.ai/families/kling Pixio's largest family, and the most complete video toolkit here — text-to-video, image-to-video, video-to-video, audio-driven video, and lipsync all in one place. Pixio's largest family, and the most complete video toolkit here — text-to-video, image-to-video, video-to-video, audio-driven video, and lipsync all in one place. If you are doing video work and have no strong preference, this is the default worth trying first. ## Which one to pick Match the entry point to what you have: a prompt, a still, or existing footage. Within each, the cheaper models are for testing motion and framing, the pricier ones for the take you keep. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (25) · text-to-video (16) · video-to-video (13) · audio-driven video (4) · image editing (3) · lipsync (2) · text-to-image (2) · audio transformation (1) · audio generation (1) **Price range:** 5–300 credits per run **67 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------------------------------------- | ------------- | ------- | ------------- | | Kling Create Voice | Audio → audio | 5 | All plans | | Kling AI Avatar Standard | Audio → video | 60 | All plans | | Kling AI Avatar V2 Standard | Audio → video | 60 | All plans | | Kling AI Avatar Pro | Audio → video | 120 | All plans | | Kling AI Avatar V2 Pro | Audio → video | 120 | All plans | | Kling O1 Image | Image → image | 10 | Free on Maker | | Kling O3 Image to Image | Image → image | 10 | Free on Maker | | Kling V3 Image to Image | Image → image | 10 | Free on Maker | | Kling 3.0 Motion Control Maker | Image → video | 35 | Free on Maker | | Kling V2.5 Turbo Standard Image to Video | Image → video | 45 | All plans | | Kling V2.1 Standard Image to Video | Image → video | 60 | All plans | | Kling V2.5 Turbo Pro Image to Video | Image → video | 75 | All plans | | Kling O1 Standard First/Last Frame to Video | Image → video | 90 | All plans | | Kling O1 Standard Reference Image to Video | Image → video | 90 | All plans | | Kling O3 Standard Image to Video | Image → video | 90 | All plans | | Kling O3 Standard Reference to Video | Image → video | 90 | All plans | | Kling Effects | Image → video | 100 | Free on Maker | | Kling V2.1 Pro Image to Video | Image → video | 105 | All plans | | Kling o1 First/Last Frame to Video | Image → video | 120 | All plans | | Kling o1 Reference Image to Video | Image → video | 120 | All plans | | Kling O3 Pro Image to Video | Image → video | 120 | All plans | | Kling O3 Pro Reference to Video | Image → video | 120 | All plans | | Kling V3 Turbo Standard Image to Video | Image → video | 120 | All plans | | Kling V3 Standard Image to Video | Image → video | 135 | All plans | | Kling 2.6 I2V Maker | Image → video | 140 | Free on Maker | | Kling V2.6 Pro Image to Video | Image → video | 150 | All plans | | Kling V3 Turbo Pro Image to Video | Image → video | 150 | All plans | | Kling V3 Pro Image to Video | Image → video | 175 | All plans | | Kling O3 4K Image to Video | Image → video | 280 | All plans | | Kling O3 4K Reference to Video | Image → video | 280 | All plans | | Kling V3 4K Image to Video | Image → video | 280 | All plans | | Kling V2 Master Image to Video | Image → video | 295 | All plans | | Kling V2.1 Master Image to Video | Image → video | 295 | All plans | | Kling LipSync Audio to Video | Lipsync | 15 | All plans | | Kling LipSync Text to Video | Lipsync | 15 | All plans | | Kling TTS | Text → audio | 10 | All plans | | Kling O3 Text to Image | Text → image | 10 | Free on Maker | | Kling V3 Text to Image | Text → image | 10 | Free on Maker | | Kling V2.5 Turbo Pro Text to Video | Text → video | 75 | All plans | | Kling O3 Standard Text to Video | Text → video | 90 | All plans | | Kling 3.0 Standard Maker | Text → video | 105 | Free on Maker | | Kling O3 Pro Text to Video | Text → video | 120 | All plans | | Kling V3 Turbo Standard Text to Video | Text → video | 120 | All plans | | Kling V3 Standard Text to Video | Text → video | 135 | All plans | | Kling 2.6 Pro Maker | Text → video | 140 | Free on Maker | | Kling 3.0 Pro Maker | Text → video | 140 | Free on Maker | | Kling V2.6 Pro Text to Video | Text → video | 150 | All plans | | Kling V3 Turbo Pro Text to Video | Text → video | 150 | All plans | | Kling V3 Pro Text to Video | Text → video | 175 | All plans | | Kling O3 4K Maker | Text → video | 280 | Free on Maker | | Kling O3 4K Text to Video | Text → video | 280 | All plans | | Kling V3 4K Text to Video | Text → video | 280 | All plans | | Kling V2 Master Text to Video | Text → video | 295 | All plans | | Kling V2.1 Master Text to Video | Text → video | 295 | All plans | | Kling Video to Audio | Video → video | 10 | All plans | | Kling O1 Standard Reference Video to Video | Video → video | 135 | All plans | | Kling O3 Standard Video to Video (Reference) | Video → video | 135 | All plans | | Kling V2.6 Standard Motion Control | Video → video | 150 | All plans | | Kling o1 Reference Video to Video | Video → video | 175 | All plans | | Kling O3 Pro Video to Video (Reference) | Video → video | 175 | All plans | | Kling O1 Standard Edit Video | Video → video | 180 | All plans | | Kling O3 Standard Video to Video (Edit) | Video → video | 180 | All plans | | Kling V3 Standard Motion Control | Video → video | 180 | All plans | | Kling V2.6 Pro Motion Control | Video → video | 250 | All plans | | Kling o1 Edit Video | Video → video | 300 | All plans | | Kling O3 Pro Video to Video (Edit) | Video → video | 300 | All plans | | Kling V3 Pro Motion Control | Video → video | 300 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # LTX Source: https://support.myapps.ai/families/ltx Video family with unusual depth in video-to-video and audio-driven video, alongside standard text-to-video and image-to-video.. Video family with unusual depth in video-to-video and audio-driven video, alongside standard text-to-video and image-to-video. ## Which one to pick Reach for this family when you already have footage or audio to work from, rather than generating from scratch. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** video-to-video (12) · text-to-video (7) · audio-driven video (4) · image-to-video (4) · audio generation (1) **Price range:** 5–260 credits per run **28 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ---------------------------------------------- | ------------- | ------------------- | ------------- | | LTX 2 Audio to Video | Audio → video | 15 | All plans | | LTX 2.3 22B Distilled Audio-to-Video LoRA | Audio → video | 60 | All plans | | LTX 2.3 22B Audio-to-Video LoRA | Audio → video | 75 | All plans | | LTX 2.3 Audio-to-Video | Audio → video | 75 | All plans | | LTX 2.3 Fast Image-to-Video | Image → video | 30 | All plans | | LTX 2.3 22B Distilled Image-to-Video LoRA | Image → video | 50 | All plans | | LTX 2.3 22B Image-to-Video LoRA | Image → video | 60 | All plans | | LTX 2.3 Image-to-Video | Image → video | 60 | All plans | | LTX 2.3 Quality Text to Audio | Text → audio | 36 | All plans | | LTX 2 Fast | Text → video | 5 | All plans | | LTX 2 Pro | Text → video | 10 | All plans | | LTX 2.3 Fast Text-to-Video | Text → video | 30 | All plans | | LTX 2.3 22B Distilled Text-to-Video LoRA | Text → video | 50 | All plans | | LTX 2.3 22B Text-to-Video LoRA | Text → video | 60 | All plans | | LTX 2.3 Pro | Text → video | 60 | All plans | | LTX 2.3 Text-to-Video | Text → video | 60 | All plans | | LTX 2 Fast Extend Video | Video → video | 5 / 24 num\_frames | Free on Maker | | LTX 2 Extend Video | Video → video | 10 / 24 num\_frames | All plans | | LTX 2 Video to Video | Video → video | 10 / 24 num\_frames | All plans | | LTX 2.3 22B Distilled Reference Video-to-Video | Video → video | 14 | All plans | | LTX 2.3 22B Reference Video-to-Video | Video → video | 18 | All plans | | LTX 2.3 22B Distilled Video-to-Video LoRA | Video → video | 60 | All plans | | LTX 2.3 22B Extend Video LoRA | Video → video | 75 | All plans | | LTX 2.3 22B Video-to-Video LoRA | Video → video | 75 | All plans | | LTX 2.3 Extend Video | Video → video | 75 | All plans | | LTX 2.3 Retake Video | Video → video | 75 | All plans | | LTX 2 Retake | Video → video | 90 | All plans | | LTX 2.3 Reframe | Video → video | 260 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Luma Source: https://support.myapps.ai/families/luma Video-first family with notable video-to-video depth, plus text-to-video, image-to-video, and image models.. Video-first family with notable video-to-video depth, plus text-to-video, image-to-video, and image models. ## Which one to pick Strong for restyling and transforming existing clips rather than only generating new ones. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** video-to-video (5) · text-to-video (3) · image-to-video (3) · text-to-image (2) · image editing (2) **Price range:** 7–150 credits per run **15 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------------------- | ------------- | ------- | ------------- | | Uni-1 Edit | Image → image | 7 | All plans | | Uni-1 Max Edit | Image → image | 16 | All plans | | Ray 2 Flash Image to Video | Image → video | 14 | Free on Maker | | Ray 2 Image to Video | Image → video | 40 | Free on Maker | | Ray 3.2 Image to Video | Image → video | 72 | All plans | | Uni-1 Text to Image | Text → image | 7 | All plans | | Uni-1 Max Text to Image | Text → image | 15 | All plans | | Ray 2 Flash Text to Video | Text → video | 14 | Free on Maker | | Ray 2 Text to Video | Text → video | 40 | Free on Maker | | Ray 3.2 Text to Video | Text → video | 72 | All plans | | Ray 2 Flash Modify | Video → video | 50 | Free on Maker | | Ray 2 Flash Reframe | Video → video | 75 | Free on Maker | | Ray 2 Modify | Video → video | 100 | Free on Maker | | Ray 3.2 Video to Video | Video → video | 104 | All plans | | Ray 2 Reframe | Video → video | 150 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Magnific Source: https://support.myapps.ai/families/magnific Upscaling and enhancement — image editing models focused on adding resolution and detail.. Upscaling and enhancement — image editing models focused on adding resolution and detail. ## Which one to pick The finishing step. Generate at a comfortable size, then upscale the one you keep. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image editing (4) · text-to-image (1) **Price range:** 6–15 credits per run **5 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------------- | ------------- | ------- | ------------- | | Upscaler (Creative) | Image → image | 6 | All plans | | Upscaler (Precision) | Image → image | 8 | All plans | | Relight | Image → image | 15 | Free on Maker | | Restyle | Image → image | 15 | Free on Maker | | Mystic | Text → image | 10 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Meshy Source: https://support.myapps.ai/families/meshy 3D family covering generation plus rigging and animation — the only place in Pixio that rigs and animates a mesh.. 3D family covering generation plus **rigging and animation** — the only place in Pixio that rigs and animates a mesh. ## Which one to pick Come here when a 3D model needs to move, not just exist. Generate or refine the mesh first, then rig, then animate. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-3D (2) · image-to-3D (2) · 3D rigging (1) · 3D animation (1) **Price range:** 10–50 credits per run **6 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------------- | ------------ | ------- | ------------- | | Animations | 3D animation | 50 | Free on Maker | | Rigging | 3D rigging | 50 | Free on Maker | | Image to 3D | Image → 3D | 50 | Free on Maker | | Multi-Image to 3D | Image → 3D | 50 | Free on Maker | | Text to 3D (Preview) | Text → 3D | 10 | Free on Maker | | Text to 3D (Refine) | Text → 3D | 30 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # NVIDIA Source: https://support.myapps.ai/families/nvidia Text-to-image models.. Text-to-image models. ## Which one to pick Compare against Flux and Stable Diffusion on the same prompt. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (5) **Price range:** 1–4 credits per run **5 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------- | ------------ | ------- | ------------------------------------------ | | Sana Base | Text → image | 1 | Free on Maker, Premium, Pro, Starter | | Sana Sprint | Text → image | 1 | Free on Free, Maker, Premium, Pro, Starter | | Sana v1.5 | Text → image | 1 | Free on Maker, Premium, Pro, Starter | | Sana v1.5 fast | Text → image | 1 | Free on Maker, Premium, Pro, Starter | | Cosmos 3 Super | Text → image | 4 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # OpenAI Source: https://support.myapps.ai/families/openai Image family split evenly between text-to-image and image editing.. Image family split evenly between text-to-image and image editing. ## Which one to pick Good prompt comprehension — you can describe what you want in plain language rather than stacking modifiers. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (4) · image editing (4) **Price range:** 3–26 credits per run **8 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ----------------------- | ------------- | ------- | ------------- | | GPT Image 1 (mini) Edit | Image → image | 4 | Free on Maker | | GPT Image 1 Edit | Image → image | 8 | Free on Maker | | GPT Image 1.5 Edit | Image → image | 8 | Free on Maker | | GPT Image 2 Edit | Image → image | 26 | Free on Maker | | GPT Image 1 (mini) | Text → image | 3 | Free on Maker | | GPT Image 1 | Text → image | 6 | Free on Maker | | GPT Image 1.5 | Text → image | 6 | Free on Maker | | GPT Image 2 | Text → image | 18 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Pika Source: https://support.myapps.ai/families/pika Compact video family — image-to-video and text-to-video.. Compact video family — image-to-video and text-to-video. ## Which one to pick Animating a still is the stronger path here. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (5) · text-to-video (3) **Price range:** 20–50 credits per run **8 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ---------------------------- | ------------- | ------- | ------------- | | Pika v2 Turbo Image to Video | Image → video | 20 | Free on Maker | | Pika v2.2 Image to Video | Image → video | 20 | Free on Maker | | Pika v2.2 Pikascenes | Image → video | 20 | Free on Maker | | Pika v2.1 Image to Video | Image → video | 40 | Free on Maker | | Pika v1.5 Pikaffects | Image → video | 50 | Free on Maker | | Pika v2 Turbo Text to Video | Text → video | 20 | Free on Maker | | Pika v2.2 Text to Video | Text → video | 20 | Free on Maker | | Pika v2.1 Text to Video | Text → video | 40 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Pixio Source: https://support.myapps.ai/families/pixio Pixio's own models — video-to-video, frame extraction, and image and audio utilities. Pixio's own models — video-to-video, frame extraction, and image and audio utilities. These power several of the chained actions in Generate. ## Which one to pick Mostly utility work: pulling frames out of video, transforming clips, and cleanup rather than headline generation. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** video-to-video (4) · frame extraction (3) · audio generation (1) · image editing (1) · text-to-image (1) **Price range:** 0–20 credits per run **10 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | --------------------- | ------------- | ------- | ------------------------------------ | | Pixio Image Edit | Image → image | 0 | Free on Maker, Premium, Pro, Starter | | Pixio Music | Text → audio | 5 | Free on Maker, Premium, Pro, Starter | | Ideogram V4 | Text → image | 20 | Free on Maker | | Extract First Frame | Video → image | 0 | Free on Maker, Premium, Pro, Starter | | Extract Frame at Time | Video → image | 0 | Free on Maker, Premium, Pro, Starter | | Extract Last Frame | Video → image | 0 | Free on Maker, Premium, Pro, Starter | | Add Audio to Video | Video → video | 0 | Free on Maker, Premium, Pro, Starter | | Black Bar Crop | Video → video | 0 | Free on Maker, Premium, Pro, Starter | | Merge Videos | Video → video | 0 | Free on Maker, Premium, Pro, Starter | | Trim Video or Audio | Video → video | 0 | Free on Maker, Premium, Pro, Starter | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # PixVerse Source: https://support.myapps.ai/families/pixverse Heavily weighted toward image-to-video — the largest image-to-video selection in Pixio — plus text-to-video, video-to-video, and lipsync.. Heavily weighted toward image-to-video — the largest image-to-video selection in Pixio — plus text-to-video, video-to-video, and lipsync. ## Which one to pick Best used the chained way: generate a still you like, then animate it. That gives you control over composition that a text prompt alone will not. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (16) · text-to-video (7) · video-to-video (5) · lipsync (1) **Price range:** 15–83 credits per run **29 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | --------------------------------- | ------------- | ------- | ------------- | | PixVerse v4.5 Image to Video Fast | Image → video | 15 | All plans | | PixVerse v4.5 Effects | Image → video | 30 | All plans | | PixVerse v4.5 Image to Video | Image → video | 30 | All plans | | PixVerse v4.5 Transition | Image → video | 30 | All plans | | PixVerse v5 Effects | Image → video | 30 | All plans | | PixVerse v5 Image to Video | Image → video | 30 | All plans | | PixVerse v5 Transition | Image → video | 30 | All plans | | PixVerse v5.5 Effects | Image → video | 30 | All plans | | PixVerse v5.5 Image to Video | Image → video | 30 | All plans | | PixVerse v5.5 Transition | Image → video | 30 | All plans | | PixVerse v5.6 Image to Video | Image → video | 60 | All plans | | PixVerse v5.6 Transition | Image → video | 60 | All plans | | PixVerse v6 Image to Video | Image → video | 66 | Free on Maker | | PixVerse v6 Transition | Image → video | 66 | Free on Maker | | PixVerse C1 Image to Video | Image → video | 83 | All plans | | PixVerse C1 Transition | Image → video | 83 | All plans | | PixVerse Lipsync | Lipsync | 30 | All plans | | PixVerse v4.5 Text to Video Fast | Text → video | 15 | All plans | | PixVerse v4.5 Text to Video | Text → video | 30 | All plans | | PixVerse v5 Text to Video | Text → video | 30 | All plans | | PixVerse v5.5 Text to Video | Text → video | 30 | All plans | | PixVerse v5.6 Text to Video | Text → video | 60 | All plans | | PixVerse v6 Text to Video | Text → video | 66 | Free on Maker | | PixVerse C1 Text to Video | Text → video | 83 | All plans | | PixVerse Extend Fast | Video → video | 15 | All plans | | PixVerse Extend | Video → video | 30 | All plans | | PixVerse Sound Effects | Video → video | 30 | All plans | | PixVerse Video Swap | Video → video | 30 | All plans | | PixVerse v6 Extend | Video → video | 66 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Maker Mode models Source: https://support.myapps.ai/families/prodia A large set of image and video models bundled for Maker Mode, spanning text-to-image, image editing, video, and vector output. A large set of image and video models bundled for Maker Mode, spanning text-to-image, image editing, video, and vector output. These are the models Maker is built around. ## Which one to pick On Maker, cost stops being the deciding factor for the included models — pick on output quality instead. Check the Maker page for which models are included versus still metered. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (12) · image editing (9) · image-to-video (6) · text-to-video (6) · vector output (2) · video-to-video (1) **Price range:** 3–400 credits per run **36 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | --------------------------------- | ------------- | ------- | ------------- | | Seedream 4 Edit | Image → image | 3 | Free on Maker | | Flux 2 Dev Edit | Image → image | 5 | Free on Maker | | Seedream 4.5 Edit | Image → image | 5 | Free on Maker | | Seedream 5.0 Lite Edit | Image → image | 5 | Free on Maker | | Gemini 3.1 Flash Edit | Image → image | 7 | Free on Maker | | Flux 2 Flex Edit | Image → image | 10 | Free on Maker | | Flux 2 Pro Edit | Image → image | 10 | Free on Maker | | Gemini 3 Pro Edit | Image → image | 12 | Free on Maker | | Flux 2 Max Edit | Image → image | 15 | Free on Maker | | Seedance Pro Image to Video | Image → video | 10 | Free on Maker | | Seedance Pro Turbo Image to Video | Image → video | 10 | Free on Maker | | Wan 2.2 Lightning Image to Video | Image → video | 100 | Free on Maker | | Wan 2.7 Image to Video | Image → video | 100 | Free on Maker | | Veo Fast Image to Video | Image → video | 200 | Free on Maker | | Veo Image to Video | Image → video | 400 | Free on Maker | | Seedream 4 Text to Image | Text → image | 3 | Free on Maker | | Wan 2.7 Text to Image | Text → image | 3 | Free on Maker | | Flux 2 Dev Text to Image | Text → image | 5 | Free on Maker | | Seedream 4.5 Text to Image | Text → image | 5 | Free on Maker | | Seedream 5.0 Lite Text to Image | Text → image | 5 | Free on Maker | | Gemini 3.1 Flash Text to Image | Text → image | 7 | Free on Maker | | Flux 2 Flex Text to Image | Text → image | 10 | Free on Maker | | Flux 2 Pro Text to Image | Text → image | 10 | Free on Maker | | Recraft V4 Text to Image | Text → image | 10 | Free on Maker | | Gemini 3 Pro Text to Image | Text → image | 12 | Free on Maker | | Flux 2 Max Text to Image | Text → image | 15 | Free on Maker | | Recraft V4 Pro Text to Image | Text → image | 40 | Free on Maker | | Recraft V4 SVG | Text → SVG | 25 | Free on Maker | | Recraft V4 Pro SVG | Text → SVG | 50 | Free on Maker | | Seedance Pro Text to Video | Text → video | 10 | Free on Maker | | Seedance Pro Turbo Text to Video | Text → video | 10 | Free on Maker | | Wan 2.2 Lightning Text to Video | Text → video | 100 | Free on Maker | | Wan 2.7 Text to Video | Text → video | 100 | Free on Maker | | Veo Fast Text to Video | Text → video | 200 | Free on Maker | | Veo Text to Video | Text → video | 400 | Free on Maker | | Wan 2.7 Video to Video | Video → video | 100 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Qwen Source: https://support.myapps.ai/families/qwen Image family weighted toward editing rather than generation — most models here transform an image you already have.. Image family weighted toward editing rather than generation — most models here transform an image you already have. ## Which one to pick Bring an image. If you need to create one first, generate elsewhere and edit here. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image editing (7) · text-to-image (5) **Price range:** 2–10 credits per run **12 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ------------------------- | ------------- | ------- | ------------- | | Qwen Image 3 Edit | Image → image | 2 | Free on Maker | | Qwen Image Edit | Image → image | 3 | Free on Maker | | Qwen-Image Edit Plus | Image → image | 4 | Free on Maker | | Qwen-Image Edit Plus Lora | Image → image | 4 | Free on Maker | | Qwen Image 2 Edit | Image → image | 5 | Free on Maker | | Qwen Image 2 Pro Edit | Image → image | 10 | All plans | | Qwen Image Max Edit | Image → image | 10 | Free on Maker | | Qwen Image | Text → image | 2 | Free on Maker | | Qwen Image 3 | Text → image | 2 | Free on Maker | | Qwen Image 2 | Text → image | 5 | Free on Maker | | Qwen Image 2 Pro | Text → image | 10 | All plans | | Qwen Image Max | Text → image | 10 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Recraft Source: https://support.myapps.ai/families/recraft Design-oriented image family, and the main source of vector/SVG output in Pixio alongside text-to-image and editing.. Design-oriented image family, and the main source of **vector/SVG output** in Pixio alongside text-to-image and editing. ## Which one to pick The reason to come here is vector output — logos, icons, and anything that has to scale cleanly. Raster models elsewhere cannot do that. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (8) · vector output (4) · image editing (1) **Price range:** 2–60 credits per run **13 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ------------------------- | ------------- | ------- | ------------- | | Recraft V3 Vectorize | Image → image | 2 | Free on Maker | | Recraft V2 | Text → image | 4 | Free on Maker | | Recraft V3 | Text → image | 8 | Free on Maker | | Recraft V4 | Text → image | 10 | Free on Maker | | Recraft V4.1 | Text → image | 15 | All plans | | Recraft V4.1 Utility | Text → image | 15 | All plans | | Recraft V4 Pro | Text → image | 40 | All plans | | Recraft V4.1 Pro | Text → image | 50 | All plans | | Recraft V4.1 Utility Pro | Text → image | 50 | All plans | | Recraft V4 (Vector) | Text → SVG | 25 | All plans | | Recraft V4.1 (Vector) | Text → SVG | 25 | All plans | | Recraft V4 Pro (Vector) | Text → SVG | 50 | All plans | | Recraft V4.1 Pro (Vector) | Text → SVG | 60 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Reve Source: https://support.myapps.ai/families/reve Image family weighted toward editing, with text-to-image alongside.. Image family weighted toward editing, with text-to-image alongside. ## Which one to pick Bring an image to transform. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image editing (6) · text-to-image (2) **Price range:** 2–33 credits per run **8 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | --------------- | ------------- | ------- | ------------- | | Reve Fast Edit | Image → image | 2 | Free on Maker | | Reve Fast Remix | Image → image | 2 | Free on Maker | | Reve Edit | Image → image | 5 | Free on Maker | | Reve Remix | Image → image | 5 | Free on Maker | | Reve 2.1 Edit | Image → image | 33 | All plans | | Reve 2.1 Remix | Image → image | 33 | All plans | | Reve | Text → image | 5 | Free on Maker | | Reve 2.1 | Text → image | 33 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Runway Source: https://support.myapps.ai/families/runway Video family with image models alongside — image-to-video, video-to-video, text-to-image, and editing.. Video family with image models alongside — image-to-video, video-to-video, text-to-image, and editing. ## Which one to pick Familiar if you have used Runway elsewhere. Image-to-video is the deepest part here. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (3) · video-to-video (2) · text-to-image (1) · image editing (1) **Price range:** 5–60 credits per run **7 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ---------------------------------- | ------------- | ------- | ------------- | | Runway Gen-4 (References to Image) | Image → image | 5 | Free on Maker | | Gen-4 (Image to Video) | Image → video | 50 | Free on Maker | | Runway Gen-4 Turbo | Image → video | 50 | Free on Maker | | Runway Gen-4.5 | Image → video | 60 | All plans | | Runway Gen-4 (Text to Image) | Text → image | 5 | Free on Maker | | Runway Act Two (Character) | Video → video | 50 | Free on Maker | | Runway Gen-4 Aleph | Video → video | 50 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # SongCraft Source: https://support.myapps.ai/families/songcraft Music generation — full songs from a prompt.. Music generation — full songs from a prompt. ## Which one to pick For songs and music beds. For speech or voice cloning, see ElevenLabs. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** audio generation (2) **Price range:** 80 credits per run **2 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ----------- | ------------ | ------- | ------------- | | Extend Song | Text → audio | 80 | Free on Maker | | Songcraft | Text → audio | 80 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Stable Diffusion Source: https://support.myapps.ai/families/stable-diffusion The long-standing open image family — text-to-image, editing, and inpainting — with a few audio models in the mix.. The long-standing open image family — text-to-image, editing, and inpainting — with a few audio models in the mix. ## Which one to pick Familiar behavior if you have used Stable Diffusion elsewhere; prompt habits carry over. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-image (5) · image editing (2) · audio transformation (2) · inpainting (1) · audio generation (1) **Price range:** 0–18 credits per run **11 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | --------------------------------- | ------------- | ------- | ------------------------------------------ | | Stable Audio 2.5 - Audio to Audio | Audio → audio | 18 | All plans | | Stable Audio 2.5 - Inpaint | Audio → audio | 18 | All plans | | SDXL Image to Image | Image → image | 0 | Free on Free, Maker, Premium, Pro, Starter | | SD 3 Medium Image to Image | Image → image | 4 | Free on Maker, Premium, Pro, Starter | | SDXL Inpainting | Inpainting | 0 | Free on Free, Maker, Premium, Pro, Starter | | Stable Audio 2.5 - Text to Audio | Text → audio | 15 | All plans | | SD 1.5 | Text → image | 0 | Free on Free, Maker, Premium, Pro, Starter | | SDXL | Text → image | 1 | Free on Free, Maker, Premium, Pro, Starter | | SD 3.5 Medium | Text → image | 2 | Free on Maker, Premium, Pro, Starter | | SD 3 Medium | Text → image | 4 | Free on Maker, Premium, Pro, Starter | | SD 3.5 Large | Text → image | 7 | Free on Maker, Premium, Pro, Starter | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Sync Source: https://support.myapps.ai/families/sync Dedicated lipsync — matching mouth movement to an audio track.. Dedicated lipsync — matching mouth movement to an audio track. ## Which one to pick Specialists rather than generalists. Bring a face and an audio track. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** lipsync (6) **Price range:** 2–20 credits per run **6 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------------------- | ------- | ------- | ------------- | | Lipsync 1.9 | Lipsync | 2 | Free on Maker | | Lipsync 2.0 | Lipsync | 5 | All plans | | Lipsync 2.0 Pro | Lipsync | 10 | All plans | | Lipsync 3.0 | Lipsync | 15 | All plans | | Lipsync 3.0 Image to Video | Lipsync | 15 | All plans | | React 1 | Lipsync | 20 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Tripo Source: https://support.myapps.ai/families/tripo Dedicated 3D family — text-to-3D and image-to-3D, evenly split.. Dedicated 3D family — text-to-3D and image-to-3D, evenly split. ## Which one to pick Start from an image when you have one; results are more predictable than from a prompt alone. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-3D (5) · text-to-3D (4) **Price range:** 20–60 credits per run **9 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ----------------------------------- | ---------- | ------- | ------------- | | Image to 3D · Tripo v3.1 (H3.1) | Image → 3D | 20 | All plans | | Multiview to 3D · Tripo v3.1 (H3.1) | Image → 3D | 20 | All plans | | Image to 3D | Image → 3D | 50 | Free on Maker | | Multiview to 3D | Image → 3D | 50 | Free on Maker | | Image to 3D · Tripo P1 | Image → 3D | 60 | All plans | | Text to 3D · Tripo v3.1 (H3.1) | Text → 3D | 20 | All plans | | Refine Model | Text → 3D | 50 | Free on Maker | | Text to 3D | Text → 3D | 50 | Free on Maker | | Text to 3D · Tripo P1 | Text → 3D | 60 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # More video models Source: https://support.myapps.ai/families/video-roundup Smaller video families, including Maker-oriented and lipsync-adjacent options.. Smaller video families, including Maker-oriented and lipsync-adjacent options. ## Which one to pick Useful as alternates when a larger family is degraded or does not fit. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** text-to-video (4) · image-to-video (3) · lipsync (3) · text-to-image (1) · video-to-video (1) **Price range:** 5–50 credits per run **12 models in Pixio.** Credits are per run unless a unit is shown. | Family | Model | Does | Credits | Access | | ------------ | --------------------- | ------------- | ------- | ------------- | | Runway Maker | Gen-4 Maker | Image → video | 14 | Free on Maker | | Runway Maker | Gen-4 Turbo Maker | Image → video | 14 | Free on Maker | | Wan | Wan 2.2 Animate Maker | Image → video | 32 | Free on Maker | | Veed | Fabric 1.0 | Lipsync | 10 | All plans | | Veed | Fabric 1.0 Fast | Lipsync | 10 | All plans | | Veed | VEED Lipsync V2 | Lipsync | 50 | All plans | | PixCraft | PixCraft Image | Text → image | 5 | All plans | | Runway Maker | HappyHorse 1.0 Maker | Text → video | 21 | Free on Maker | | Wan | Wan 2.6 Flash Maker | Text → video | 35 | Free on Maker | | Runway Maker | Gen-4.5 Maker | Text → video | 42 | Free on Maker | | Pruna AI | P-Video | Text → video | 44 | All plans | | Veed | VEED Subtitles | Video → video | 50 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Vidu Source: https://support.myapps.ai/families/vidu Compact video family weighted toward image-to-video, with text-to-video and video-to-video options.. Compact video family weighted toward image-to-video, with text-to-video and video-to-video options. ## Which one to pick Animate a still you have already approved rather than generating video from a cold prompt. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (7) · text-to-video (3) · video-to-video (1) **Price range:** 30–100 credits per run **11 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | -------------------------------- | ------------- | ------- | ------------- | | Vidu Q1 Image to Video | Image → video | 40 | Free on Maker | | Vidu Q1 Reference to Video | Image → video | 40 | Free on Maker | | Vidu Q1 Start-End to Video | Image → video | 40 | Free on Maker | | Vidu Q2 Image to Video Turbo | Image → video | 40 | Free on Maker | | Vidu Q2 Image to Video Pro | Image → video | 50 | Free on Maker | | Vidu Q3 Image to Video | Image → video | 100 | All plans | | Vidu Q3 Reference to Video (Mix) | Image → video | 100 | All plans | | Vidu Q2 Text to Video | Text → video | 30 | Free on Maker | | Vidu Q1 Text to Video | Text → video | 40 | Free on Maker | | Vidu Q3 Text to Video | Text → video | 100 | All plans | | Vidu Q2 Video Extension Pro | Video → video | 50 | Free on Maker | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Wan Source: https://support.myapps.ai/families/wan Balanced family spanning image-to-video, video-to-video, text-to-image, text-to-video, and image editing — an even spread rather than one specialty.. Balanced family spanning image-to-video, video-to-video, text-to-image, text-to-video, and image editing — an even spread rather than one specialty. ## Which one to pick Useful when a project needs both stills and motion in a consistent look, since you can stay inside one family. Every input has a tooltip in [Generate](/generate) — read those before assuming a setting exists on a given model. ## Models **Capabilities:** image-to-video (8) · text-to-video (6) · video-to-video (6) · text-to-image (5) · image editing (2) **Price range:** 3–198 credits per run **27 models in Pixio.** Credits are per run unless a unit is shown. | Model | Does | Credits | Access | | ---------------------------------- | ------------- | ------- | ------------------------------------ | | WAN 2.6 Image to Image | Image → image | 5 | Free on Maker | | WAN 2.7 Pro Edit | Image → image | 10 | All plans | | WAN 2.5 Image to Video | Image → video | 10 | Free on Maker | | WAN v2.2 Image to Video | Image → video | 10 | Free on Maker | | WAN Effects | Image → video | 35 | Free on Maker | | WAN 2.6 Image to Video | Image → video | 50 | All plans | | WAN 2.7 Image to Video | Image → video | 50 | All plans | | WAN 2.7 Reference to Video | Image → video | 50 | All plans | | Happy Horse 1.1 Image to Video | Image → video | 115 | All plans | | Happy Horse 1.1 Reference to Video | Image → video | 115 | All plans | | WAN v2.2 Text to Image | Text → image | 3 | Free on Maker, Premium, Pro, Starter | | WAN 2.5 Text to Image | Text → image | 5 | Free on Maker, Premium, Pro, Starter | | WAN 2.6 Text to Image | Text → image | 5 | Free on Maker | | WAN 2.7 Text to Image | Text → image | 5 | Free on Maker | | WAN 2.7 Pro Text to Image | Text → image | 10 | All plans | | WAN 2.5 Text to Video | Text → video | 10 | Free on Maker | | WAN v2.2 Text to Video | Text → video | 10 | Free on Maker | | WAN 2.6 Text to Video | Text → video | 50 | All plans | | WAN 2.7 Text to Video | Text → video | 50 | All plans | | Happy Horse 1.1 Text to Video | Text → video | 115 | All plans | | Happy Horse 1.0 | Text → video | 198 | All plans | | WAN v2.2 Video to Video | Video → video | 10 | Free on Maker | | WAN Animate Move | Video → video | 20 | Free on Maker | | WAN Animate Replace | Video → video | 20 | Free on Maker | | WAN VACE Video Edit | Video → video | 20 | Free on Maker | | WAN 2.7 Edit Video | Video → video | 50 | All plans | | WAN 2.6 Reference to Video | Video → video | 75 | All plans | *Capabilities, price range, and table generated from the Pixio model catalog.* ## Credits and access Costs are per run unless the table shows a unit. Access follows your plan — see [Credits & Plans](/credits). Models marked free on a plan run without drawing down your balance. Test at the cheapest settings the model allows before committing. For video, generate a still first and chain **Create Video** from it — see [chained actions](/generate#chained-actions). ## Related * [Models](/models) — how the catalog works * [Generate](/generate) — where you run these * [Credits & Plans](/credits) — what things cost * [Model status](https://beta.pixio.myapps.ai/status/models) — if one is misbehaving # Audio FAQ Source: https://support.myapps.ai/faq/audio Common questions about music, voice, and sound in Pixio ## Choosing a model [SongCraft](/families/songcraft) generates full songs from a prompt. [MiniMax](/families/minimax) has notable audio depth too, and [Mureka](/families/audio-avatar-roundup) is worth comparing. [ElevenLabs](/families/elevenlabs) is the default for text-to-speech and voice work. Filter for **audio-to-audio**. [ElevenLabs](/families/elevenlabs) and [MiniMax](/families/minimax) both offer it. Look at **text-to-audio** models — they cover effects as well as music. Read the model descriptions, since capability names don't distinguish the two. Audio quality is subjective in a way image quality isn't. Generate the same prompt on two or three models and listen. There's no substitute for that. ## Voice and video together Two steps. Generate the voice track (**Generate Voice** [chained action](/generate#chained-actions) or [ElevenLabs](/families/elevenlabs)), then sync it — **Add Lipsync** on an image, **Lipsync Video** on a clip, or [Sync](/families/sync) directly. Yes — **audio-to-video** models generate video from an audio track. [Kling](/families/kling) and [LTX](/families/ltx) both have them. [HeyGen](/families/heygen) handles presenter video directly, rather than assembling voice and lipsync yourself. ## Costs Often **per second**, like video — so a long track costs proportionally more. The cost is shown before you submit. See [Credits & Plans](/credits). Generate a short clip to check the voice or musical style before committing to full length. Style is audible in a few seconds. ## Practical [My Assets](/assets) → Generated, alongside images and video. Download from there or from the preview. Yes — upload it in the form or pick it from your assets, for lipsync, audio-to-video, or audio transformation. Yes — the [Video Editor](/video-editor) assembles clips and audio. Generate the pieces, then arrange them there. Check [status](/status) — audio is its own service group. Failed runs aren't charged. ## Related * [ElevenLabs](/families/elevenlabs) · [SongCraft](/families/songcraft) · [MiniMax](/families/minimax) · [Sync](/families/sync) * [Generate](/generate) · [Credits & Plans](/credits) # Image FAQ Source: https://support.myapps.ai/faq/image Common questions about generating and editing images in Pixio ## Choosing a model [Flux](/families/flux) — the widest range in Pixio, strong prompt adherence, and the best LoRA support. Compare against [Stable Diffusion](/families/stable-diffusion) and [OpenAI](/families/openai) on the same prompt before settling. Try [Ideogram](/families/ideogram) — it's the family best known for legible text in images. You want vector output, not a raster image. [Recraft](/families/recraft) is the main source of SVG output in Pixio. Filter for **image-to-image** or **inpainting**. [Qwen](/families/qwen), [Reve](/families/reve), and [Flux](/families/flux) all lean editing. For a change described in words, the Edit Image action from [Grok](/families/grok) works well. Same capability doesn't mean same controls. Two text-to-image models can accept completely different inputs. Check the tooltips on each input in [Generate](/generate) rather than assuming a setting exists. ## Quality and prompts Usually the model, not the prompt. Prompt adherence varies a lot between families — try the same prompt on a [Flux](/families/flux) model before rewriting it. If one model consistently ignores you, it's the wrong model for that job. Fix the **seed** if the model exposes one — same seed and same prompt gives you a repeatable starting point. For a consistent subject or character across many images, train a [LoRA](/advanced/model-training). That's what LoRA training is for. Train on 15–30 varied images of the subject, then trigger it with its trigger word. See [Model Training](/advanced/model-training). Don't regenerate — upscale. Use the **Upscale** or **Image Upscale** [chained action](/generate#chained-actions), or [Magnific](/families/magnific) directly. ## Costs Depends on the model — from a couple of credits up to a few dozen. The cost is shown in the picker and again before you submit. See [Credits & Plans](/credits). Draft on a cheap model at low resolution to get composition and prompt right, then rerun the keeper on a better model. Iterating on the expensive model is the most common way people burn a balance. Successful generations cost credits even if the output isn't what you wanted. Only **failed** runs are free. That's the reason to test cheap first. ## Inputs and outputs Upload it in the generation form, or pick an existing file from your assets. Uploads are reusable — they land in the Uploads tab of [My Assets](/assets). [My Assets](/assets) → Generated, and in the history column in [Generate](/generate). The prompt is in [Prompt History](/library/prompt-history). Yes — the **Remove BG** chained action, or a background-removal model in [3D & utility](/families/3d-and-utility). ## Related * [Models](/models) — how to navigate the catalog * [Generate](/generate) — the studio in detail * [Academy](/academy) — model-specific lessons # Getting support Source: https://support.myapps.ai/faq/support Check status, find the right docs, then reach a human ## Start here [Status](/status) reports image, video, audio, and workflow availability separately, plus per-model status. Most "it stopped working" questions are answered there — and **failed runs aren't charged**, so an incident costs you nothing but time. [Image](/faq/image) · [Video](/faq/video) · [Audio](/faq/audio) If status is clean and the FAQ doesn't cover it, report it — with the details below. ## Common answers Check [status](/status) first, including per-model status — one model can be degraded while its whole category looks healthy. Failed runs aren't charged. Trying a different model in the same capability is usually the fastest unblock. Upgrade or buy a credit pack on [Billing](/credits). Packs add permanent credits on top of your monthly allowance. Access is per plan. The picker marks what your plan can't run rather than hiding it. See [Credits & Plans](/credits). Check [Analytics](/advanced/analytics) — it breaks generations down by model and family. Video and audio often bill per second, so duration matters more than run count. [My Assets](/assets) holds everything, and [Prompt History](/library/prompt-history) holds every prompt. If you were in a team account when you made it, switch to that account — assets belong to the account that created them. A team account is one shared pool and any member can spend it. There's no per-member limit. See [Accounts & Teams](/billing/account). See [API keys](/integrations/api-keys) — a `401` usually means the key is disabled or the header is malformed; a `403` means it lacks the scope. ## Reporting a problem Include these and it gets diagnosed on the first reply: * **What you ran** — the model and the surface (Generate, Agent, Boards…) * **When** — roughly, so it can be matched against incidents * **What happened** versus what you expected * **Whether it repeats**, or happened once * **Whether other models work** for the same capability Never include an API key or integration token in a support message or screenshot. If you've already sent one, disable it in **Integrations** immediately. ## Other products Your own ComfyUI workflows as an API — separate product, separate API. Start with its [troubleshooting](/comfyui/troubleshooting) page. The iPad drawing and tracing app — its own setup and support. API keys, the Claude skill, and what you can automate. A separate product with its own docs in this site. # Video FAQ Source: https://support.myapps.ai/faq/video Common questions about generating video in Pixio Video is the most expensive output in Pixio. Almost every question below has the same underlying answer: **decide on cheap stills first, commit to video once.** ## Choosing a model [Kling](/families/kling) — the largest and most complete video family here, covering text-to-video, image-to-video, video-to-video, audio-driven video, and lipsync. Then compare [ByteDance](/families/bytedance), [Google](/families/google), and [PixVerse](/families/pixverse). **Image-to-video, almost always.** Generate a still until the composition is right, then animate it. Text-to-video makes you pay video prices to discover the framing was wrong. Use the **Create Video** [chained action](/generate#chained-actions). Filter for **video-to-video**. [LTX](/families/ltx) and [Luma](/families/luma) have the most depth there. [HeyGen](/families/heygen) for avatars and presenters. To sync an existing face to an audio track, [Sync](/families/sync) is the specialist. That's [Video Agent](/create/video-agent) — segment generation with character memory and a stitched export. Plan it in [Cinema Studio](/create/cinema-studio) first. ## Control Prompting a camera angle is unreliable. [CamView](/advanced/camview) lets you block the shot in 3D — place objects and figures, position the camera, keyframe the move — so framing is geometry instead of a description. Blocking is free; you only pay to render. [Video Agent](/create/video-agent) carries character memory across segments. For a specific recurring subject, train a [LoRA](/advanced/model-training) and use it to generate consistent stills, then animate those. [Storyboards](/create/storyboards) in Cinema Studio — frames with their own descriptions, shot language, and per-frame references. Changing a storyboard frame is free. Duration limits are per model and shown in the form. For anything beyond one clip, generate segments and assemble them in the [Video Editor](/video-editor) or via Video Agent. ## Costs Many video models bill **per second**, and cost also scales with resolution. A short test clip and a final render can differ by an order of magnitude. In order of effect: 1. Get the first frame right as a cheap **image**, then chain into video 2. Test at the **shortest duration** the model allows 3. Test at **lower resolution** before the final run 4. Block the camera in [CamView](/advanced/camview) — free — instead of re-rolling prompts No. Failed generations are never charged. ## Finishing The [Video Editor](/video-editor) — a real timeline. Note that **rendering is app-only**; it can't be triggered through the API. Yes — **Upscale Video**, **Caption Video**, **Lipsync Video**, **Edit Video**, and **Remove BG** are all [chained actions](/generate#chained-actions) on a finished video. Check [status](/status) — video is reported as its own service group, and per-model status can show one model degraded while the rest are fine. Failed runs aren't charged, so don't retry into a known incident. ## Related * [Kling](/families/kling) · [ByteDance](/families/bytedance) · [Google](/families/google) · [PixVerse](/families/pixverse) * [Video Agent](/create/video-agent) · [Cinema Studio](/create/cinema-studio) · [CamView](/advanced/camview) * [Credits & Plans](/credits) # Generate Source: https://support.myapps.ai/generate The main studio for images, video, audio, and 3D **Generate** is the core of Pixio. You pick a model, fill in its inputs, and run it. Results land in your history, your assets, and your prompt history automatically. ## Layout Model list and the generation form. The form changes to match whichever model you select. Full-size view of the current result. Images, video, audio, and 3D all preview in place. Every generation, newest first, with chained actions on each one. Scrolls to load more. ## The top bar Alongside **Home**, **Create**, **Analytics**, **My Assets** and **Prompt History**, three controls sit to the right of the navigation: | Control | What it does | | -------------------- | ----------------------------------------------------------------------------------------------------------------------------------- | | **Pixio Legacy** | Brings your work across from Pixio v1 — **Sync Prompts** and **Sync Trained LoRAs**. Run it once if you're coming from the old app. | | **Prompt Optimizer** | Rewrites a prompt for better results. Available on **paid plans**; on Free you're prompted to upgrade. | | **Your plan** | Shows your current subscription (e.g. *Maker Monthly*) | Your credit balance and profile menu sit at the far right. **Pixio Legacy** is the migration path from v1. Prompts and trained LoRAs don't move across on their own — sync them and they appear in [Prompt History](/library/prompt-history) and in the LoRA picker. ## Choosing a model Models are grouped by the family that makes them, and each entry shows: * **What it does** — text-to-image, image-to-video, video-to-video, lipsync, text-to-3D, and so on * **Credit cost** — per run, or per second for models billed by duration * **Access** — whether your plan can run it Over 500 models are available. See [Models](/models) for how to narrow that down. The picker has **Search models…**, a type filter (**All types**), a sort (**Default**), and a **Maker** toggle to show only Maker-eligible models. Below that, **Favorites** first, then **All Models** grouped by family with a count per family. Credits show as an exact number for fixed-price models and with a `~` for models whose cost varies with duration or resolution. ### Favorites Tap the star on any model to favorite it. Favorites are saved to your account and pinned in their own section at the top of the picker, so the handful of models you actually use stay one click away. ### Plan access Your plan controls which models you can run. Models outside your plan are marked in the list. Some models are free on specific plans, so they don't draw down your balance at all. Details in [Credits & Plans](/credits). ## The form Each model exposes its own inputs — prompt, image size, quality, output format, duration, resolution, seed, style, guidance. **Every input has a Field information tooltip** explaining what it does and what range it accepts, so you don't have to guess. At the bottom of the form, above the Generate button, you'll see: | | | | ---------------------------------- | ------------------------------------------------------------------------------------------------ | | **Estimated cost** | What this run will cost, updating as you change settings | | **Maker full-discount runs (24h)** | On Maker, how many full-discount runs you have left in the rolling 24-hour window for this model | The Maker discount counter is per rolling 24 hours, not per calendar day, and applies to models in a capped group. When it's used up the model still runs — it just draws on credits. See [Maker Mode](/advanced/maker-mode). ### Using your own images and video Models that take a reference image, source video, or audio track let you either: * **Upload a file** directly from the form, or * **Pick from your assets** — anything you've generated or uploaded before Uploads are stored against your account and show up in the **Uploads** tab of [My Assets](/assets), ready to reuse. ### LoRAs If you've trained a custom LoRA, models that support it expose a LoRA picker in the form. Trigger your LoRA by including its trigger word in the prompt. LoRA training requires Premium or above. ## Running a generation The run is created immediately and appears in the right column as **processing**. **Estimated cost** shows what it will cost before you commit. You can navigate away, start another generation, or close the tab — the job keeps running. Images are usually seconds; video can take several minutes. On success the output appears in the preview and history. On failure you are **not charged**. ### Not enough credits If your balance won't cover the run, Pixio tells you before starting it and points you at [Billing](/credits) to top up or upgrade. Nothing is queued and nothing is charged. ## Chained actions This is the part worth learning. Hover any finished item in the right column and you can push it straight into the next model — no downloading, no re-uploading. **From an image** | Action | What it does | | ------------------------------- | ----------------------------------------- | | **Regenerate** | Run again with the same prompt and inputs | | **Create Video** | Animate the still into a video | | **Next Frame** | Continue the sequence from this frame | | **Create 3D Model** | Turn the image into a 3D mesh | | **Upscale** / **Image Upscale** | Increase resolution | | **Remove BG** | Cut out the background | | **Photo Restoration** | Repair and clean up a photo | | **Edit Image** | Make a targeted edit from an instruction | | **Add Lipsync** | Drive a face from audio | | **Generate Voice** | Produce a voice track | **From a video** | Action | What it does | | ----------------- | ---------------------------------------- | | **Upscale Video** | Increase resolution | | **Caption Video** | Burn in captions | | **Lipsync Video** | Sync mouth movement to audio | | **Edit Video** | Make a targeted edit from an instruction | | **Remove BG** | Cut the background out of the clip | | **No Watermark** | Produce a clean version | **From a 3D model** **Refine 3D model** improves mesh quality on models that support refinement. The cheapest way to get good video: generate several inexpensive images, pick the best one, then chain **Create Video** from it. Iterating at image prices instead of video prices saves an enormous number of credits. ## Your results Every finished generation is available in three places automatically: * **History** — the right column, newest first * **[My Assets](/assets)** — your full library, with uploads on a second tab * **Prompt History** — every prompt you've used, searchable and reusable From the preview or history you can **download** a result, **publish** it to the public gallery, **unpublish** it later, or **delete** it. Generations belong to the account that made them. On a team account, everyone on the team shares the same assets and the same credit pool. # Agents & plugins Source: https://support.myapps.ai/integrations/agents-and-plugins Let an external app act on your Pixio account, and revoke it when you're done Some tools connect to Pixio without you ever pasting a key. Instead the app asks for access, you approve it in the browser while signed in, and it receives its own session. ## How authorization works A plugin or agent starts an authorization session and opens Pixio in your browser. You land on the Pixio authorization page. Because you're signed in, Pixio knows which account is granting access. Approve, and that session is linked to the requesting app. The app polls until authorization completes, then holds its own session — no key copied by hand, no key sitting in a config file. From then on it acts on your account, spending your credits and writing into your assets. Nothing is granted until you approve in a signed-in browser. An app can start the request, but it can't complete it on your behalf. ## What an authorized app can do The same things you can, subject to your plan: * Generate media and spend your credits * Read and write your assets * Read your credit balance and plan * Run workflows you've saved It is **your** account. There's no separate sandbox, and no separate balance. ## Before you approve Only approve apps you actually recognize. An authorized app can spend every credit you have. If an authorization page appears when you didn't just try to connect something, close it and don't approve. Worth asking: * **Did I start this?** An authorization page you didn't trigger is a red flag. * **Do I trust it with credits?** Approval is account access, not a read-only peek. * **Can I revoke it?** Yes — see below. Approval is reversible. ## Revoking access Signing out of the granting session ends the app's access. If you're unsure what's connected or want a clean slate: 1. Change your password, which invalidates existing sessions. 2. Review your integration keys in **Integrations** and disable any you don't recognize. If you want per-integration control you can turn off individually, prefer an [API key](/integrations/api-keys) over plugin authorization. Keys are named, scoped, and independently disableable. ## Choosing between the two | | Plugin authorization | [API key](/integrations/api-keys) | | ------------------- | ------------------------------- | --------------------------------- | | Setup | Approve in browser | Create and paste a key | | Best for | Tools built to connect to Pixio | Your own code and scripts | | Scoping | Account access | Choose specific areas | | Revoke | End the session | Disable that one key | | Good for CI/servers | No | Yes | # API keys Source: https://support.myapps.ai/integrations/api-keys Create, scope, and disable integration keys An **integration key** lets code outside Pixio act on your account. Create one per place you use it, so you can disable that one without breaking the others. ## Creating a key Click **Integrations** in the sidebar. Name the key after where it will be used — `Production`, `Local dev`, `Zapier`. The name is how you'll recognize it later. Select which areas the key may touch (below). Grant only what the integration needs. Copy the key immediately and store it somewhere safe. Treat it like a password. ## What a key can reach Access is grouped, so you can hand out narrow keys: | Scope | What it allows | | ---------------------------------- | -------------------------------------------------------- | | **Media & workflows** | Generate, upload, manage assets, and run saved workflows | | **Creative projects** | Author Boards, Canvas, Cinema, and Cam View scenes | | **Agent & editor projects** | Persist agent workspaces and apply timeline operations | | **Characters, prompts & training** | Manage reusable context and inspect training progress | A key that only needs to generate images doesn't need project or training access. Narrow scopes limit the damage if a key leaks. ## Using a key Send it as a Bearer token: ```bash theme={null} curl https://beta.pixio.myapps.ai/api/v1/models \ -H "Authorization: Bearer YOUR_INTEGRATION_KEY" ``` Everything a key does is billed to the same credit balance and constrained by the same plan as your account in the app. ## Disabling and re-enabling Keys are **disabled**, not deleted — so the record of what existed stays intact. * **Disable key** stops it working immediately. Any integration using it starts failing on the next request. * Disabled keys are hidden by default; toggle **Show disabled** to see them. * A disabled key can be **enabled** again if you disabled it by mistake. If a key leaks, disable it first and ask questions after. A leaked key can spend every credit on your account. ## Good practice * **One key per environment.** Never share a key between production and local development. * **Server-side only.** A key in client-side JavaScript or a mobile app binary is a public key. * **Environment variables**, never source control. Add your `.env` to `.gitignore`. * **Rotate on staff change.** Create the replacement, switch over, then disable the old one — in that order, so nothing breaks mid-cutover. * **Name honestly.** `test2` tells you nothing in six months. ## Troubleshooting The key is wrong, disabled, or missing from the header. Check the header is exactly `Authorization: Bearer `, then confirm the key is still Active in **Integrations**. The key is valid but lacks the scope for that operation. Check its scopes against the table above. The account is out of credits — the same wall you'd hit in the app. Top up on [Billing](/credits). Access is per plan, not per surface. If your plan can't run it in the UI, a key can't either. Key storage isn't configured on that environment. This affects self-hosted or preview deployments, not the hosted app. # Claude skill Source: https://support.myapps.ai/integrations/claude-skill Generate Pixio media from a conversation with Claude The **Pixio skill** lets Claude create media on your account — ask for an image, a video, or a voice track in conversation and Claude runs the generation through Pixio. Installation steps, supported commands, and the current version live in the repository README. ## What it's good for * **Generating without leaving your workflow** — you're already in a conversation, you don't switch tabs * **Letting Claude pick the model** — describe the outcome instead of choosing from 500 models * **Iterating in language** — "same thing, wider shot, warmer" rather than re-filling a form ## What you need Generations cost the same credits as they do in the app. Free accounts work — the free balance is enough to try it. Create one in **Integrations**. See [API keys](/integrations/api-keys) — give it **Media & workflows** scope at minimum. Follow the README in the repository above. Keep the key in an environment variable, not pasted into a chat. Never paste an integration key into a conversation. Configure it where the skill reads it from the environment. Anything typed into a chat is stored in that conversation's history. ## What it can reach Whatever the key's scopes allow — generation, assets, and saved workflows for a standard key. It's the same account, so: * Results appear in **My Assets** exactly as if you'd generated them in the app * Prompts are saved in **Prompt History** * Costs come out of the same balance * Your plan's model access applies unchanged ## Limits worth knowing * **Long video takes minutes.** A conversation isn't the ideal place to wait for a slow render — start it, then check the app. * **Some surfaces are app-only.** Timeline renders, training jobs, and autonomous Video Agent runs aren't reachable from any integration. See [what you can automate](/integrations/what-you-can-automate). * **Failed generations are free**, but a run that succeeds and simply isn't what you wanted still costs credits. Iterate on cheap image models first. The skill is one way in, not a separate product. Anything it makes is a normal Pixio generation you can open, download, chain, or publish from the app. # Integrations Source: https://support.myapps.ai/integrations/overview Connect Pixio to the tools you already work in Everything you can do in the Pixio app, you can do from somewhere else — your own code, an AI assistant, or an automation. One integration key unlocks all of it. Create a key, choose what it can reach, and disable it when you're done. Generate images, video, and audio directly from a conversation with Claude. Let an external app request access to your Pixio account, and revoke it later. The honest list — what's supported, and what isn't. ## Which one do I want? | If you want to… | Use | | ------------------------------------------ | ----------------------------------------------------------------------------------- | | Generate from your own script or backend | An **API key** plus the [live API reference](https://beta.pixio.myapps.ai/api-docs) | | Create from a chat with Claude | The **[Pixio skill](/integrations/claude-skill)** | | Let a third-party tool act on your account | **[Plugin authorization](/integrations/agents-and-plugins)** | | Run a chain you've already built in Pixio | An API key — saved **workflows** are callable | | Build ComfyUI graphs and call them | **[Pixio ComfyUI](/comfyui/overview)**, a separate product with its own API | ## How it works In the app, open **Integrations** and create an integration key. Name it after where it'll be used. Keys are scoped. Grant only the areas the integration actually needs. Send the key as a Bearer token. Every request is billed to the same credit balance as the app. Start a generation, then poll it until it finishes. Results land in your assets exactly as if you'd made them in the app. ## Same account, same credits An integration isn't a separate world. Anything created through a key: * **Costs the same credits** as it would in the app * **Respects your plan** — a model your plan can't run through the UI can't run through a key either * **Appears in your assets** and generation history * **Counts toward the same limits** Failed generations aren't charged, through the API just as in the app. ## Reference The complete endpoint reference — every model, operation, and parameter, with copyable request examples — lives in the app at **[/api-docs](https://beta.pixio.myapps.ai/api-docs)**. It's generated from the catalog and the API's own specification, so it can't drift from the real behavior. Treat an integration key like a password. It can spend your credits. Never commit one to a repository or ship it in client-side code — keep it server-side, in an environment variable. # What you can automate Source: https://support.myapps.ai/integrations/what-you-can-automate Exactly which parts of Pixio are reachable from an integration — and which aren't Most of Pixio is scriptable, but not all of it. This page is the honest boundary, so you find out here rather than halfway through a build. ## Supported Start generations, estimate their credit cost before running, and check status and history. List, upload, import, fetch, download, and delete generated and uploaded media. List models, inspect their parameters, and optimize prompts. Read credit balance, ledger, plan, and API limits. List workflows already saved in Pixio and run them. Full project CRUD and operations for Boards, spatial canvas, and Canvas designs. Storyboard CRUD with prompt-to-storyboard direction; Cam View scene CRUD with prompt-to-keyed-scene direction. Persist chat and Video Agent projects; timeline project CRUD with validated editor operations. Locked character CRUD, prompt history, and model-training status. ## Not supported These are deliberately app-only. Don't design around them. | Not available | Why it matters | | ---------------------------------------------------------------------- | -------------------------------------------------------------------------- | | **Running Agent chat tools** through the internal streaming chat route | You can persist a conversation, but you can't drive the agent's tool calls | | **Cam View controller pairing** and live sensor streaming | Scene setup is scriptable; the live phone-as-camera link isn't | | **Headless timeline renders**, exports, and render progress | You can build and edit a timeline; you can't trigger the final render | | **Autonomous Video Agent orchestration** and final assembly | Project state is scriptable; the end-to-end autonomous run isn't | | **Starting or cancelling training jobs** | You can read training *status*, but jobs are started in the app | The pattern: **authoring and reading are open, long-running orchestration and live device links are not.** You can construct the work, then finish it in the app. ## Planning around the gaps Generate clips through the API and assemble them yourself, or build the timeline via the API and do the final render in the app. Headless render isn't exposed. Not possible today — training starts in the app. You *can* poll training status and react when one finishes. Build it on the generation endpoints directly rather than the internal chat route. You get the same models without depending on an unsupported surface. Build it once as a **workflow** in the app, then call it by ID. Cheaper than reimplementing the chain in your own code, and it stays in sync when you edit it. ## Reference Every supported operation has its own page with parameters and a live playground, generated from the API's own specifications: | Section | Covers | Endpoints | | ---------------------- | ----------------------------------------------------------------------------------------------------------------------------- | --------- | | **Pixio Integrations** | Generation, assets, models, credits, saved workflows | 29 | | **Pixio Project API** | Boards, Canvas, Cinema storyboards, Cam View scenes, chat, Video Agent, editor projects, characters, prompt library, training | 55 | `GET /capabilities` returns the supported surfaces programmatically, so an integration can check rather than assume. # Pixio Source: https://support.myapps.ai/introduction Generate images, video, audio, and 3D from one workspace Pixio is an AI creation workspace. Pick a model, describe what you want, and generate — **images, video, audio, music, voice, and 3D models**, all from the same place, all billed from one credit balance. Over **500 models** are available in the app, from the families you already know: Flux, Kling, Veo, Seedance, Wan, Luma, Hailuo, PixVerse, LTX, Recraft, Ideogram, ElevenLabs, Hunyuan, Meshy, and many more. Sign up, spend your first credits, and find your result. Five minutes. The main studio — how the three panels work and how to chain outputs. What a generation costs, what each plan includes, and what happens when you run out. How the catalog works — capabilities, plan access, and picking the right model. ## What you can make Text-to-image, image editing, inpainting, upscaling, background removal, restoration, and vector/SVG output. Text-to-video, image-to-video, video-to-video restyling, interpolation, captions, upscaling, and lipsync. Music generation, sound effects, text-to-speech, and voice cloning. Text-to-3D, image-to-3D, mesh refinement, rigging, and animation. ## Ways to create Generate is the core, but it isn't the only way in. | Surface | Use it when | | ----------------- | --------------------------------------------------------------------------------------- | | **Generate** | You know the model you want and you want direct control over every input. | | **Pix Agent** | You’d rather describe the goal in chat and let Pixio choose and run models. | | **Video Agent** | You want a multi-segment video built and stitched from one brief. | | **Cinema Studio** | You're planning a video as a storyboard before generating any frames. | | **Canvas** | You want to draw, trace, or composite in layers and generate from what's on the canvas. | | **Video Editor** | You have clips and need a real timeline — trimming, captions, export. | | **Workflows** | You run the same multi-step chain repeatedly and want it saved. | | **Boards** | You want generations organized into projects and shareable. | ## Getting started Sign up at [beta.pixio.myapps.ai](https://beta.pixio.myapps.ai). You get **50 free credits** that never expire. Open **Generate**, pick a model, write a prompt, and run it. Cheap image models cost only a few credits, so your free balance goes a long way while you're learning. Every result lands in **My Assets** and in your generation history, and every prompt is saved in **Prompt History**. Working with a team? A team account has its own shared credit balance and shared assets, separate from your personal account. See [Teams & Account](/billing/account). # Gallery & Showcase Source: https://support.myapps.ai/library/gallery-vs-showcase The two public surfaces, and which one your work belongs on Pixio has two public surfaces. They're easy to confuse, so: | | [Gallery](https://beta.pixio.myapps.ai/gallery) | [Showcase](https://beta.pixio.myapps.ai/showcase) | | ------------- | ----------------------------------------------- | ------------------------------------------------- | | What's on it | Images, video, audio, and 3D made with Pixio | AI-generated videos and films | | Scope | Everything the community publishes | Curated cinema | | Feel | Browse what's possible | Watch finished work | | Creator pages | — | Yes, per creator | **Gallery** is the broad public feed of published work across every output type. **Showcase** is the video and film surface, with its own creator pages. ## Publishing You publish from your own work — in [Generate](/generate) or [My Assets](/assets), choose to publish a finished generation to the public gallery. Anything succeeded, in your history or assets. It becomes visible on the public gallery. Publication is reversible. Unpublishing removes it from public view. ## Before you publish Publishing makes work **publicly visible**. Don't publish client work, anything under NDA, images of people who haven't agreed, or anything you'd want back. Unpublishing removes it from Pixio, but it can't retrieve what someone already saw or saved. Worth checking: * **Do I have the rights?** Trained on someone's likeness or a client's product? Get agreement first. * **Is anything identifying in it?** Check the image itself, not just the prompt. * **Would I be happy attached to this?** Published work is associated with you. ## Publishing vs sharing Two different jobs: * **Publish** when you want the work *found* — gallery and showcase are discoverable * **[Share a link](/library/share-links)** when you want it *seen by specific people* — not discoverable, but anyone with the URL can view ## Getting inspired The gallery is a research tool as much as a display case. Browsing it shows what the current models are genuinely capable of — usually more informative than reading a model description. ## Related * [My Assets](/assets) — publish and unpublish * [Share links](/library/share-links) — private sharing * [Generate](/generate) — publish straight from a result # Prompt History Source: https://support.myapps.ai/library/prompt-history Every prompt you've used, by category, ready to reuse **Prompt History** keeps every prompt you've generated with, organized by category — image, video, audio, 3D. It's the fastest way back to something that worked. ## Finding a prompt | Tool | What it does | | -------------- | ----------------------------------------- | | **Categories** | Filter to All, Image, Video, Audio, or 3D | | **Search** | Search across your prompts | | **Cards** | Click any card to see the full prompt | ## What you can do with one Take the text to adapt elsewhere. Send it straight back into Generate. Jump to what that prompt produced. Long prompts are truncated on the card — open to read all of it. ## Why it matters Good prompts are work. Prompt History means that work isn't lost when you close a tab: * **Reuse what worked** instead of reconstructing it from memory * **Compare** two prompts that produced different results * **Reuse across models** — a prompt that worked on one model is a strong starting point on another When something turns out well, note *why* in the prompt itself if you can. Six months later "cinematic, low angle, 35mm" is legible; a wall of modifiers isn't. Prompt history is populated automatically — every generation writes to it. Nothing to save by hand. ## Related * [My Assets](/assets) — the outputs * [Generate](/generate) — where prompts get reused * [what you can automate](/integrations/what-you-can-automate) — prompt history is readable through the API # Share links Source: https://support.myapps.ai/library/share-links Share work with people who don't have a Pixio account Share links let someone view your work without signing up. Each link carries its own token, so you're sharing one specific thing — not access to your account. ## What can be shared | You share | They see | | --------------------- | ------------------------------- | | **A generated asset** | The single image, clip, or file | | **A board** | The board and its outputs | | **A canvas** | The canvas composition | | **A CamView scene** | The scene via its VCam link | ## How it works Share from the board, canvas, or asset you want to share. Anyone with the link can view it — no account, no sign-in. A share link is read-only. Recipients can't change your work or spend your credits. ## Share links vs the gallery | | Share link | [Gallery / Showcase](/library/gallery-vs-showcase) | | ------------ | ---------------------------------- | -------------------------------------------------- | | Who finds it | Only people you send it to | Anyone browsing Pixio | | Discoverable | No | Yes | | Best for | A client, a collaborator, feedback | Showing work publicly | | Undo | Stop sharing | Unpublish | Share a link for review. Publish to the gallery when you want the work seen. ## Security **A share link is a URL that works for whoever holds it.** There's no password and no per-person access list — if the link is forwarded, the recipient can view it too. Don't share anything confidential this way. Practical guidance: * Send links to specific people, not public channels * Assume a shared link may be forwarded * For a wide audience, publish to the gallery instead — that's what it's for ## Related * [Boards](/create/boards) · [Canvas](/create/canvas) · [CamView](/advanced/camview) * [My Assets](/assets) — where individual assets live * [Gallery vs Showcase](/library/gallery-vs-showcase) — public publishing # Advanced Tutorials Source: https://support.myapps.ai/machine/advanced-tutorials Step-by-step guides for complex tasks # Advanced Tutorials These step-by-step tutorials demonstrate how to use Machine for complex, real-world tasks. Each tutorial provides detailed instructions, examples, and best practices. ## Data Analysis Workflow This tutorial walks through using Machine to analyze a dataset, generate insights, and create visualizations. Start by uploading your dataset to Machine. Supported formats include CSV, Excel, JSON, and SQL databases. ``` You: "I need to analyze this customer feedback dataset" Machine: "I can help with that. Would you like to upload the dataset now?" ``` After uploading, Machine will automatically examine the data structure and provide an initial summary. Next, clean your data to handle missing values, outliers, and inconsistencies. ``` You: "The dataset has some missing values and outliers we need to address" Machine: "I notice there are 23 missing values in the 'customer_age' column and 12 potential outliers in 'purchase_amount'. Would you like me to: 1. Remove rows with missing values 2. Impute missing values using the mean/median 3. Handle these issues another way?" ``` Machine will suggest transformations based on data quality issues it detects, but you maintain control over how to proceed. Have Machine perform initial analysis to understand patterns and relationships in the data. ``` You: "What are the main patterns you notice in this data?" Machine: "Based on my analysis, here are the key insights: - Customer satisfaction correlates strongly with response time (r=0.72) - The 25-34 age group represents 42% of your customer base - Product category A has the highest return rate at 8.3% Would you like me to generate visualizations for these insights?" ``` Machine can identify correlations, trends, clusters, and outliers automatically. Create visual representations of your data insights. ``` You: "Yes, please create visualizations for the key insights" Machine: "I'll create several visualizations to highlight these patterns. Which format would you prefer for the final report: PDF, interactive dashboard, or presentation slides?" ``` Machine will generate appropriate visualizations based on data types and relationships, including: * Bar charts for categorical comparisons * Scatter plots for correlations * Time series for trends * Heatmaps for complex relationships Finally, have Machine provide actionable recommendations based on the analysis. ``` You: "Based on this analysis, what actions should we consider taking?" Machine: "Based on the data, here are my recommendations: 1. Reduce response times for the 25-34 demographic to improve satisfaction 2. Investigate quality issues in Product Category A to address high return rates 3. Expand inventory for high-margin products showing consistent growth Would you like me to prepare a detailed action plan for any of these recommendations?" ``` Machine can help prioritize actions based on potential impact and implementation difficulty. ## Website Creation Process This tutorial demonstrates how to use Machine to build a complete website from scratch. Start by describing your website needs and goals. ``` You: "I need to create a website for my photography business" Machine: "I'd be happy to help create your photography website. To get started, could you tell me: 1. What's the primary purpose of the website? (portfolio, booking, sales) 2. How many pages do you envision? 3. Do you have any design preferences or examples of sites you like?" ``` Be as specific as possible about your requirements, but don't worry if you don't have all the details figured out—Machine will help fill in the gaps. Work with Machine to plan your website's architecture. ``` You: "I need a portfolio showcase, about page, services/pricing, and contact form" Machine: "Based on your needs, here's a proposed site structure: - Home (with featured work and introduction) - Portfolio (organized by photography categories) - About (your story and approach) - Services & Pricing - Contact Does this structure work for you, or would you like to make adjustments?" ``` Machine will suggest a logical organization based on your business type and website purpose. Develop the text and visual content for your website. ``` You: "I need help writing the content for these pages" Machine: "Let's start with the About page. Could you tell me a bit about your background, style, and approach to photography? I'll help craft that into engaging website copy." ``` For visual elements, you can: * Upload your own images * Have Machine suggest stock photos * Get guidance on creating a cohesive visual style Let Machine generate the actual website code based on your approved design and content. ``` You: "Now let's build the actual website" Machine: "I'll create your website using modern, responsive design principles. Do you have a preference for: 1. A static site (faster, simpler) 2. A content management system like WordPress (more easily editable) 3. A custom solution with specific features?" ``` Machine can generate: * HTML/CSS/JavaScript for static sites * WordPress themes and configurations * React/Next.js applications for more advanced requirements Review the website and make iterative improvements. ``` You: "Can I see how the site looks so far?" Machine: "Here's a preview of your website. You can navigate between pages to see the current design and content. What aspects would you like to adjust?" ``` Machine will provide a live preview environment where you can: * Navigate between pages * Test responsive behavior for different screen sizes * Experiment with different design elements Finally, publish your website to make it accessible online. ``` You: "The site looks great! How do I publish it?" Machine: "I can help you deploy your website. Do you already have: 1. A domain name 2. Web hosting If not, I can guide you through acquiring these and then help with the deployment process." ``` Machine can assist with: * Domain registration * Hosting setup * DNS configuration * Deployment to platforms like Netlify, Vercel, or traditional hosts ## Research Automation Example This tutorial shows how to use Machine to conduct comprehensive research on a topic. Begin by specifying your research objectives and scope. ``` You: "I need to research the impact of artificial intelligence on healthcare" Machine: "I'll help you research AI in healthcare. To focus our efforts, could you specify: 1. Any particular aspects of healthcare you're most interested in? 2. Whether you need academic sources, news articles, or both? 3. How recent the information should be? 4. The approximate length or depth of research needed?" ``` Clear research parameters help Machine deliver more relevant results. Let Machine collect relevant information from multiple sources. ``` You: "Focus on diagnostic applications from the past 3 years, using both academic and news sources" Machine: "I'll gather information on AI in healthcare diagnostics from the past 3 years. This will include academic papers, news articles, and industry reports. This will take a few moments." ``` Machine will: * Search academic databases for relevant papers * Find news articles from reputable sources * Identify industry reports and analyses * Filter for recency and relevance Have Machine analyze the information and identify key themes and findings. ``` You: "What are the main findings from your research?" Machine: "Based on my analysis of 28 academic papers and 42 news articles, here are the key findings on AI in healthcare diagnostics: 1. Diagnostic Accuracy: Studies show 15-30% improvement in early detection rates for conditions like diabetic retinopathy and lung cancer 2. Implementation Challenges: Healthcare systems face integration issues with existing workflows and EHR systems 3. Regulatory Developments: FDA has approved 18 AI-based diagnostic tools in the past 2 years 4. Emerging Trends: Multimodal AI systems combining imaging with patient history show promising results Would you like me to elaborate on any of these areas?" ``` Machine identifies patterns across sources and synthesizes information into coherent themes. Create a structured research report with citations. ``` You: "Please create a comprehensive report with these findings" Machine: "I'll prepare a detailed research report on AI in healthcare diagnostics. Would you prefer: 1. A formal academic-style report with full citations 2. A business-oriented white paper format 3. A presentation with key points and visualizations 4. Something else?" ``` Machine can generate: * Executive summaries * Detailed reports with sections and subsections * Literature reviews with proper citations * Visual presentations of key findings Verify the accuracy and completeness of the research. ``` You: "Can you verify the accuracy of the statistics cited in the report?" Machine: "I'll verify all statistics in the report. For example, the claim that 'AI systems show 15-30% improvement in early detection rates' comes from: 1. Wong et al. (2023) in JAMA Network Open - reported 16.8% improvement 2. Patel & Johnson (2022) in Nature Medicine - reported 28.3% improvement 3. Cleveland Clinic (2024) study - reported 22.5% improvement Would you like me to provide direct links to these sources or verify any other specific claims?" ``` For critical research, Machine can provide links to original sources for manual verification. ## Custom Integrations Workshop This tutorial demonstrates how to connect Machine to external services through APIs. Determine which external services you need to connect with. ``` You: "I need to integrate Machine with our company's CRM system" Machine: "I can help you integrate with your CRM. To get started: 1. Which CRM system are you using? 2. What specific actions do you want to perform? (e.g., create contacts, update deals) 3. Do you have API credentials for the CRM?" ``` Be specific about what data you need to exchange and what actions you want to automate. Set up and secure the necessary API access. ``` You: "We use HubSpot CRM and need to create contacts and update deals. I need to get API credentials" Machine: "Great, I can guide you through obtaining HubSpot API credentials: 1. Log into your HubSpot account 2. Go to Settings > Integrations > API Keys 3. Click 'Create key' and provide a description like 'Machine Integration' 4. Copy the generated key Once you have the key, we'll use it to set up the integration." ``` Machine provides step-by-step guidance for API setup while keeping security best practices in mind. Set up the connection between Machine and your external service. ``` You: "I have the API key now" Machine: "Perfect! To securely configure the HubSpot integration: 1. In Machine, go to Settings > Integrations 2. Select 'Add New Integration' 3. Choose 'HubSpot' from the list 4. Enter your API key when prompted 5. Select the specific permissions needed (contact creation, deal updates) Would you like me to guide you through testing the connection?" ``` Machine helps you configure the integration with the appropriate scope and permissions. Develop workflows that leverage the integration. ``` You: "I want to automatically create HubSpot contacts when I receive emails from new clients" Machine: "I'll help you set up that workflow. Here's how we'll approach it: 1. Create a trigger that activates when you receive emails from unknown addresses 2. Extract contact information from the email signature and content 3. Check if the contact already exists in HubSpot 4. If not, create a new contact with the extracted information 5. Send you a notification when a new contact is created Should we set this up now?" ``` Machine can design complex workflows that combine multiple services and actions. Verify that the integration works as expected and make adjustments. ``` You: "Let's test this with a sample email" Machine: "I'll help you test the integration. Please upload or forward a sample client email, and I'll simulate the workflow to show what information would be extracted and how the HubSpot contact would be created." ``` Testing helps identify potential issues before deploying the integration in a production environment. ## Additional Tutorials Learn how to create an automated workflow for processing, analyzing, and organizing documents. Configure Machine to manage your calendar, emails, and daily tasks through integrations. Build a systematic approach to creating, editing, and publishing content across multiple platforms. Transfer and transform data between different systems while maintaining integrity and structure. These tutorials are regularly updated based on user feedback and new features. If you have a suggestion for a tutorial topic, please [let us know](mailto:rcohen@mytsi.org). # External APIs in Machine Source: https://support.myapps.ai/machine/api-in-machine Connecting Machine to external services # Using External APIs with Machine Machine can connect to virtually any external API using its built-in Python and command-line capabilities. This guide explains how to use these features to expand Machine's functionality by connecting to external services. ## API Integration Capabilities Machine can execute Python code that uses the requests library to call external APIs. For direct HTTP requests, Machine supports curl commands for quick API interactions. Securely connect to APIs that require authentication using various methods. Parse and transform API responses directly within Machine. ## Basic Python API Example Here's a simple example of using Python to connect to an external API within Machine: ```python theme={null} import requests # Make a GET request to a public API response = requests.get('https://api.example.com/data') # Check if the request was successful if response.status_code == 200: # Process the JSON response data = response.json() print(f"Retrieved {len(data)} records") # You can now use this data in your workflow for item in data[:5]: # Show first 5 items print(item['name']) else: print(f"Error: {response.status_code}") ``` Simply paste this code into Machine, adjust the URL and parsing logic for your specific API, and run it. ## Using Curl Commands If you prefer using curl commands, Machine supports those as well: ```bash theme={null} curl -X GET https://api.example.com/data -H "Authorization: Bearer YOUR_TOKEN" ``` ## Authentication Methods Machine supports various authentication methods for APIs: ### API Key Authentication ```python theme={null} import requests headers = { 'X-API-Key': 'your_api_key_here' } response = requests.get('https://api.example.com/data', headers=headers) ``` ### OAuth Authentication ```python theme={null} import requests headers = { 'Authorization': 'Bearer your_oauth_token_here' } response = requests.get('https://api.example.com/data', headers=headers) ``` ### Basic Authentication ```python theme={null} import requests response = requests.get( 'https://api.example.com/data', auth=('username', 'password') ) ``` ## POST Requests with Data Sending data to an API: ```python theme={null} import requests data = { 'name': 'New Item', 'description': 'This is a new item created via API' } response = requests.post('https://api.example.com/items', json=data) if response.status_code == 201: print("Item created successfully!") print(response.json()) else: print(f"Error: {response.status_code}") ``` ## Common API Use Cases Fetch data from external sources to enhance your analysis or combine with your own data. ```python theme={null} # Example: Getting weather data import requests city = "New York" api_key = "your_weather_api_key" response = requests.get(f"https://api.weatherapi.com/v1/current.json?key={api_key}&q={city}") weather_data = response.json() print(f"Current temperature in {city}: {weather_data['current']['temp_c']}°C") ``` Use external AI services alongside Machine for specialized content generation. ```python theme={null} # Example: Generating an image description using an external API import requests image_url = "https://example.com/image.jpg" vision_api_key = "your_vision_api_key" headers = {"Authorization": f"Bearer {vision_api_key}"} data = {"image_url": image_url} response = requests.post("https://api.vision-service.com/analyze", headers=headers, json=data) description = response.json()["description"] print(f"Image description: {description}") ``` Convert data between formats using specialized APIs. ```python theme={null} # Example: Converting currency import requests amount = 100 from_currency = "USD" to_currency = "EUR" api_key = "your_exchange_api_key" response = requests.get(f"https://api.exchangerate.host/convert?from={from_currency}&to={to_currency}&amount={amount}&api_key={api_key}") result = response.json() print(f"{amount} {from_currency} = {result['result']} {to_currency}") ``` ## Best Practices When using external APIs with Machine, follow these best practices: 1. **Security**: Never hardcode sensitive credentials directly in your code. Consider using environment variables or secure storage. 2. **Rate Limiting**: Be aware of API rate limits. Add delays between requests if making multiple calls to avoid being blocked. 3. **Error Handling**: Always implement proper error handling to manage API failures gracefully. 4. **Data Validation**: Validate the data received from external APIs before using it in critical operations. 5. **Caching**: Consider caching API responses for frequently accessed data to improve performance. ## Limitations and Considerations * Machine has network connectivity to most common APIs but may have restrictions for security reasons * Some APIs may have CORS or other security restrictions that prevent direct access * For very large datasets, consider pagination or streaming techniques * Machine operates within its security sandbox, so some system-level operations might be restricted Remember that when using external APIs, you're responsible for complying with each service's terms of use and data handling requirements. # Browser Automation Source: https://support.myapps.ai/machine/browser-automation How Machine navigates the web for you # Browser Automation Machine's browser automation capabilities allow it to navigate the web on your behalf, extracting information and performing actions across websites through simple conversational commands. ## Overview Browser automation enables Machine to act as your digital navigator, visiting websites, interacting with web elements, and gathering information—all while you guide it through natural conversation. Machine can browse the internet, follow links, and navigate through complex websites. Extract structured data from websites, convert it to useful formats, and save it for analysis. Fill out forms, submit information, and interact with web applications. Automate multi-step processes across different websites and web applications. ## Key Capabilities ### Website Navigation Machine can visit websites and navigate through them based on your instructions: ``` You: "Visit example.com and go to their products page" Machine: "I'm navigating to example.com... Now I'm on their products page. I can see they have categories for electronics, clothing, and home goods. Would you like me to explore any of these categories?" ``` ### Data Extraction Machine can extract specific information or entire datasets from websites: ``` You: "Find the top 10 most popular products on the site and create a spreadsheet with their names, prices, and ratings" Machine: "I'm scanning the products page... I've identified the top 10 products based on the 'Popular' badge. I'll extract their details and create a spreadsheet for you." ``` ### Form Filling & Submission Machine can complete forms and submit information: ``` You: "Fill out their contact form to request more information about their enterprise solutions" Machine: "I'll help you fill out the contact form. What name should I use? And what email address would you like them to respond to?" ``` ### Multi-step Workflows Machine can perform complex sequences of actions across multiple websites: ``` You: "I need to compare prices for this product across three different retailers" Machine: "I'll search for this product on Amazon, Best Buy, and Walmart to compare prices. This will take a moment..." ``` ## Use Cases Machine can monitor competitors' websites, track price changes, gather product information, and compile comprehensive market reports. Collect and organize content from multiple sources, such as news articles, blog posts, or research papers on a specific topic. Extract contact information from company directories, LinkedIn profiles, or industry-specific platforms to build targeted lead lists. Set up regular checks of websites for changes to prices, inventory, content updates, or other important information. ## Permissions & Ethical Use Machine always requires your explicit permission before accessing any website or performing actions. It's designed to respect: * **Website Terms of Service**: Machine will not attempt to circumvent website restrictions or access protected content without proper authorization. * **Rate Limiting**: Machine follows appropriate request rates to avoid overloading websites. * **Data Privacy**: Information extracted from websites is handled according to our privacy policy and used only for your specific tasks. Browser automation should only be used for legitimate purposes. Do not use Machine to scrape websites that prohibit such activities, attempt to circumvent paywalls, or violate terms of service. ## Limitations While powerful, Machine's browser automation capabilities have some limitations: * **CAPTCHA & Anti-Bot Measures**: Some websites employ measures specifically designed to prevent automation. * **Dynamic Content**: Highly dynamic websites with complex JavaScript may present challenges. * **Login Requirements**: Some sites may require user authentication that Machine cannot complete without your direct involvement. * **Geographic Restrictions**: Content that's region-restricted may not be accessible. ## Getting Started with Browser Automation To use Machine's browser automation features, simply describe the web-based task you need help with. For example: * "Can you check the current price of Product X on these three websites?" * "I need information about upcoming events in my city this weekend." * "Help me find the best deals on flights from New York to London for next month." Machine will guide you through the process, asking for clarification when needed and keeping you informed of its progress. For optimal results, be as specific as possible about what information you're looking for or what actions you want Machine to take on websites. # OpenClaw vs Built-in Telegram Source: https://support.myapps.ai/machine/claw-info How the messaging gateway (OpenClaw) and Machine’s built-in Telegram work under the hood, and when to use each. # Two Ways to Connect Telegram to Machine **Good news: both paths support Telegram.** You can connect your Machine agents to Telegram either via **Machine’s built-in Telegram** (Telegram-only, full Machine experience) or via the **messaging gateway (OpenClaw)** (Telegram + WhatsApp + Discord from one gateway). They work differently. Many customers find **Machine’s built-in Telegram integration** more powerful for Telegram and group chats — see the checklist below. This page explains how each works and when to use which. *** ## Telegram: Machine built-in vs gateway — at a glance When you use **Telegram**, you can use **Machine’s built-in Telegram** (full, app-like experience) or the **gateway** (Telegram + WhatsApp + Discord). Built-in is often the better fit for Telegram and group chats. **Streaming:** Yes — typing, “Using tool…”, streamed text.\ **Group chats:** Full support; @mentions stripped; context per chat.\ **Threads:** Same as the app — one thread per Telegram chat, in your projects.\ **Files:** Add in Machine; agent has full access.\ **Computer:** Yes when needed.\ **Manage:** Machine dashboard → Telegram page.\ **Channels:** Telegram only. **Streaming:** No — one final message.\ **Group chats:** Via gateway; one reply per message.\ **Threads:** Different model; not the same as app threads.\ **Files:** Manage in **OpenClaw** / gateway UI — agents and files live there.\ **Computer:** Different path; no computer for that run.\ **Manage:** MachineClaw + Machine “Messaging channels”.\ **Channels:** Telegram + WhatsApp + Discord. **Summary:** For Telegram and group chats, **Machine’s built-in Telegram is more powerful**: same experience as the app, streaming, proper threads, one place (Machine) to add files. Use the **gateway** when you need **Telegram + WhatsApp + Discord** and are okay with no streaming and managing OpenClaw at MachineClaw. *** ## Production readiness and open source **Production-grade.** Built-in Telegram is part of Machine and is built for production. It is maintained, monitored, and designed so it will not crash under normal use. **Open source.** OpenClaw is open source. You can self-host it, inspect the code, and modify it to fit your needs. *** ## How Machine runs an agent (quick context) In Machine, when you chat in the app or in **Agent Chatrooms**, your message is saved to a **conversation thread**, the agent gets the full setup (including the computer when needed, your files, and limits), and the agent runs with **tools** and **streaming** — you see typing, “Using tool…”, and text as it’s written. **Built-in Telegram is where messages are actually forwarded** (Telegram → Machine via webhook); it uses this same experience. The **gateway** is **isolated** (no forwarding from OpenClaw to Machine; it uses the Machine API key) and uses a different path: one reply at the end, no live streaming. *** ## Built-in Telegram — how it works Built-in Telegram is **part of Machine**. You set up the bot in the Machine dashboard. Messages are **forwarded** from Telegram to Machine (via Machine’s webhook); Machine runs the **same** agent experience as in-app chat. The user messages your bot (e.g. in a private chat or group). Telegram sends an update to Machine’s webhook. Machine looks up your bot and loads your saved settings (agent, model, etc.). Machine gets or creates a **conversation thread** for this Telegram chat (and a “Chats” project if needed). Your message is saved to that thread — same as when you type in the Machine app. Machine runs the agent with the **full setup**: computer when needed, your files, tools, and limits. The agent has the same **tools**, **computer**, and context as in the Machine app. Machine sends updates to Telegram in real time: **typing** indicator, **“🔧 Using tool…”** when a tool runs, and **streamed reply text**. The user sees the same live experience as in the app. So: **one thread per Telegram chat**, **messages saved in that thread**, **full agent run** with **real-time streaming** (typing, tools, text) to Telegram. Built-in Telegram uses the **same** agents you use in Machine chat and in **Agent Chatrooms**. Full agent (tools, computer, files). Same threads and projects as in Machine. Real-time streaming. Conversation history in the same model as the app. Machine dashboard → Telegram page. Add bot, paste token, pick agent (and optional model). Machine sets the webhook and stores the credential. **Limitation:** Built-in Telegram is **Telegram only**. There is no built-in WhatsApp or Discord in this path. *** ## Messaging gateway (OpenClaw) — how it works The **gateway** is **isolated** — a separate, sandboxed environment. OpenClaw runs inside that sandbox and has **its own chat UI**. It can connect **one** agent to **Telegram, WhatsApp, and Discord** at once. When a user sends a message, OpenClaw receives it and uses the **Machine API key** to call Machine (no forwarding from OpenClaw); Machine runs the agent and returns **one** reply. The gateway sends that reply to the user. So the user does **not** get real-time streaming (no typing, no “Using tool…” in the app) — they get the final answer in one go. The user messages your bot on one of the connected platforms. OpenClaw (inside the gateway sandbox) receives it. The gateway is isolated; it does **not** forward messages. It uses the **Machine API key** to call Machine directly. Machine runs the agent; the gateway does not run your agent itself. Machine finds or creates a **thread** for this user and agent. The conversation model is **minimal** — your message is not stored in the same way as in the app. Machine starts the agent run with the message. Machine runs the agent but **without** the full app setup (no computer for this run, no file-upload path from the app). The agent still gets the message and can use tools, but the **experience and thread model** are different from the app and built-in Telegram. Machine waits for the agent to finish, then **returns a single reply** to the gateway. The gateway sends that reply to the user on Telegram/WhatsApp/Discord. The user sees **one** message — no live streaming or tool updates in the app. So: **gateway = isolated, sandboxed, own UI, multi-channel**. It uses the **Machine API key** (no forwarding from OpenClaw to Machine). Machine runs the agent and returns **one** reply. No real-time streaming to the channel; different thread and conversation model from the app. One agent on **Telegram + WhatsApp + Discord**. Manage OpenClaw and the gateway’s chat UI at **MachineClaw** ([machineclaw.myapps.ai](https://machineclaw.myapps.ai/)). Enable and link the gateway in Machine’s “Messaging channels” / OpenClaw settings. No full app-style run (no computer for this run, no file-upload flow from the app). Machine returns **one reply** to the gateway, so **no streaming** (typing, tool status) to the user. User messages are not stored in the thread the same way as in the app. The gateway is **not** a thin relay. It runs in a **sandbox** and has **its own chat UI**. You manage it at **MachineClaw** and in Machine’s gateway settings. Choose it when you need **multi-channel** and are okay with this different, non-streaming experience. *** ## Quick comparison **Flow:** Telegram → Machine → thread + message saved → full agent run → **stream** (typing, tools, text) to Telegram.\ **Threads:** Same as Machine app (one thread per chat, in your projects).\ **Streaming:** Yes — typing, “Using tool…”, streamed reply.\ **Channels:** Telegram only.\ **Best for:** Telegram-only; want the **full** Machine experience and same agents as chat/chatrooms. **Flow:** User → OpenClaw (isolated sandbox; uses Machine API key) → Machine runs agent → one reply → gateway sends to user. No forwarding from OpenClaw to Machine.\ **Threads:** Different model; user message not stored in thread like in the app.\ **Streaming:** No — user gets **one** final message.\ **Channels:** Telegram, WhatsApp, Discord.\ **Best for:** Multi-channel; okay with sandboxed gateway and MachineClaw; no need for live streaming in the app. *** ## FAQ and common questions ### If I want to add a file to my agent that I built using the gateway method, do I add it in Machine, MachineClaw, or the gateway UI? **In OpenClaw / gateway UI.** With the gateway, you manage agents and files **inside OpenClaw** — add or edit files, instructions, and knowledge in the gateway UI. You do not manage that agent in the Machine dashboard. ### Does the agent operate differently in Machine than in MachineClaw / the gateway? Do I still need to use the gateway UI? **Yes, it operates differently depending on where the user is talking to it.** * **In Machine (app or built-in Telegram):** The agent runs with the **full** pipeline — same threads, projects, tools, computer, streaming. You don’t use the gateway UI to chat; you chat in the Machine app or in Telegram (if you use built-in Telegram). * **Via the gateway (Telegram / WhatsApp / Discord through OpenClaw):** The agent still runs in Machine, but the **path** is different: the gateway sends one message and gets one reply back. So the **experience** for the user is different (no streaming in the app, one final message). The **gateway UI** (and MachineClaw) are for **setup and management** — adding tokens, pairing, checking status — not for day-to-day chatting. To **use** the agent, people just open Telegram (or WhatsApp, etc.) and message the bot; they do **not** need to open the gateway UI. So: **you do not need to use the gateway UI to use the agent.** You use the gateway UI (or MachineClaw) to set up and manage the gateway; end users chat in the messaging app. ### Important Facts 1. **Machine built-in Telegram** — Telegram-only, part of Machine, same pipeline as the app and chatrooms (streaming, threads, files). 2. **Messaging gateway (OpenClaw)** — Sandboxed, its own UI, Telegram + WhatsApp + Discord; gateway calls Machine and gets one reply per message. *** ### Where do I add files or knowledge to my agent? **Machine (Direct / built-in Telegram):** You manage in **Machine** — add or edit files, instructions, and knowledge in the Machine dashboard. **OpenClaw (gateway):** You manage inside **OpenClaw** / the gateway UI; agents and files live there. ### Where do I manage the gateway and OpenClaw? **MachineClaw** at [https://machineclaw.myapps.ai/](https://machineclaw.myapps.ai/) — for OpenClaw and the gateway’s own chat UI. In **Machine**, use the agent’s “Messaging channels” / OpenClaw settings to enable the gateway and link it to your agent. ### Does built-in Telegram use the same agents as Machine chat and chatrooms? **Yes.** Built-in Telegram uses the **same** agents and the **same** pipeline as the Machine app and **Agent Chatrooms**. One agent, one place — whether someone chats in the app, in a chatroom, or in Telegram (built-in). ### Why would I use the gateway instead of built-in Telegram? Use the **gateway** when you need **Telegram + WhatsApp + Discord** from one agent and are okay with: the gateway’s sandboxed environment, its own UI, no real-time streaming to the user (one final reply), and managing OpenClaw at MachineClaw. Use **built-in Telegram** when you only need Telegram and want the full Machine experience (streaming, same threads/agents as the app and chatrooms). ### Do I need to use the gateway UI to chat with my agent? **No.** End users chat in Telegram (or WhatsApp, etc.). You use the gateway UI and MachineClaw for **setup and management** — tokens, pairing, status — not for day-to-day chatting. ### Can I use both built-in Telegram and the gateway for the same agent? You can use **built-in Telegram** for one bot (Telegram-only, full experience) and the **gateway** for the same or another agent (multi-channel). They are separate setups. One agent can be connected via the gateway (Telegram + WhatsApp + Discord) and you can also have a different Telegram bot connected via built-in Telegram to the same or another agent. ### Why is Machine’s built-in Telegram described as “more powerful” for Telegram? For **Telegram specifically**, built-in Telegram gives you: real-time streaming (typing, “Using tool…”, streamed text), the same thread and project model as the app (so conversations are first-class in Machine), full support for group chats and @mentions, and one place (Machine) to add files and manage the bot. The gateway, when used for Telegram, still runs the agent in Machine but returns one reply per message and uses a different thread model; there’s no streaming in the app. So for “Telegram only” use cases, built-in is usually the better fit. ### How do group chats work with built-in Telegram? Machine’s built-in Telegram supports group chats. You can configure the bot (e.g. in BotFather) so it sees all messages or only when @mentioned. Machine strips @mentions from the text before sending to the agent and can add context (who is speaking, that it’s a group). One thread per group chat; same flow as private chats. ### How do group chats work with the gateway? The gateway receives messages from Telegram groups. It uses the **Machine API key** to call Machine (no forwarding from OpenClaw to Machine); Machine runs the agent and returns one reply; the gateway sends that reply to the group. There’s no real-time streaming in the group — users see one final message. ### Where is my bot token stored for built-in Telegram? In **Machine** — in the credential store (e.g. encrypted). Machine sets the Telegram webhook and keeps the token so it can send replies. You never put the token in MachineClaw or the gateway for built-in Telegram. ### Where is my bot token stored when I use the gateway? For the **gateway**, the token is configured in the gateway (e.g. via Machine’s “Messaging channels” or OpenClaw config in the sandbox). MachineClaw and the gateway UI are where you manage that. The agent itself still lives in Machine; the gateway uses the token to receive and send messages on Telegram (or WhatsApp, Discord). ### I only need Telegram. Which should I use? Use **Machine’s built-in Telegram**. You get the full Machine experience (streaming, same threads as the app, group chats, one place to add files). No gateway, no MachineClaw needed for that bot. ### I need Telegram and WhatsApp (and maybe Discord). Which should I use? Use the **messaging gateway (OpenClaw)**. It’s the path that connects one agent to Telegram + WhatsApp + Discord. You’ll set up and manage the gateway at MachineClaw and in Machine’s “Messaging channels” settings. ### Does the agent have access to tools and computer in both paths? **Built-in Telegram:** Yes — full pipeline, so tools and computer (sandbox) when needed, same as the app.\ **Gateway:** The agent in Machine can still use tools, but the run doesn’t get the full app setup (e.g. no computer for that run). The reply is sent back as one message; the user doesn’t see “Using tool…” in the messaging app. ### Can I see the conversation history for gateway-connected chats in Machine? The gateway uses a different thread and conversation model; user messages aren’t stored in the thread the same way. So you won’t see the same per-message history in Machine as for in-app or built-in Telegram chats. The agent still runs and replies, but the conversation model is different. ### What is MachineClaw? **MachineClaw** ([machineclaw.myapps.ai](https://machineclaw.myapps.ai/)) is the UI for managing **OpenClaw** and the messaging gateway. You use it to configure channels, pairing, and the gateway’s own chat interface. It is separate from the Machine app; Machine is where your agents, chatrooms, and built-in Telegram bots live. ### What are Agent Chatrooms? **Agent Chatrooms** are a Machine feature: shared spaces where multiple agents (and optionally users) collaborate. They use the same agents and threads as the rest of Machine. Built-in Telegram uses those same agents; the gateway does not use chatrooms — it’s a separate path that calls the agent via the gateway API. ### Summary: where do I do what? | I want to… | Where | | -------------------------------------------------- | ---------------------------------------------------------------- | | Add a file or knowledge (built-in Telegram) | **Machine** (dashboard, that agent) | | Add a file or knowledge (gateway / OpenClaw) | **OpenClaw** / gateway UI (agents and files live in the gateway) | | Connect a Telegram-only bot with full experience | **Machine** → Telegram page (built-in) | | Connect one agent to Telegram + WhatsApp + Discord | **Machine** “Messaging channels” + **MachineClaw** (gateway) | | Manage the gateway / OpenClaw | **MachineClaw** + Machine “Messaging channels” | | Chat with my agent as a user | **Machine app**, or **Telegram/WhatsApp** (no gateway UI needed) | | Set up or fix the gateway (tokens, pairing) | **MachineClaw** or gateway UI / Machine “Messaging channels” | # Command-Line Source: https://support.myapps.ai/machine/command-line System automation through conversation # Command-Line Functionality Machine can execute command-line operations on your behalf, enabling powerful system automation through simple conversations. This capability allows you to perform complex technical tasks without needing to remember specific command syntax. ## Overview The command-line functionality serves as a bridge between natural language and system-level operations, allowing you to: * Run terminal commands without memorizing syntax * Execute scripts and programs * Automate system tasks * Install and configure software * Manage files and directories at the system level Command-line operations can modify your system and files. Machine always asks for confirmation before executing potentially destructive or high-impact commands. ## Available Operations ### File System Operations Machine can help you manage files and directories through commands like: ```bash theme={null} # List directory contents ls -la ~/Documents # Find files by pattern find ~/Documents -name "*.pdf" -mtime -7 # Create directories mkdir -p ~/Projects/new-project/src # Move or rename files mv ~/Downloads/document.pdf ~/Documents/important.pdf ``` ### System Information Retrieve information about your system: ```bash theme={null} # System information uname -a systeminfo # Disk usage df -h du -sh ~/Documents # Process monitoring ps aux | grep "chrome" top ``` ### Network Operations Perform network diagnostics and operations: ```bash theme={null} # Check connectivity ping google.com # Network interface information ifconfig ipconfig /all # DNS lookups nslookup example.com dig example.com ``` ### Package Management Install and manage software packages: ```bash theme={null} # For Debian/Ubuntu apt update && apt install package-name # For Red Hat/Fedora dnf install package-name # For macOS brew install package-name # For Windows choco install package-name ``` ### Process Management Control running processes: ```bash theme={null} # Start a process start program.exe open application.app # Stop a process pkill process-name taskkill /IM program.exe ``` ## Using Command-Line Functionality ### Natural Language Requests You don't need to know exact commands. Simply describe what you want to accomplish: ``` You: "I need to find all PDF files in my Downloads folder from the last week" Machine: "I can help with that. I'll use the 'find' command to locate PDF files modified in the last 7 days. Would you like me to run: find ~/Downloads -name '*.pdf' -mtime -7" ``` ### Script Creation Machine can create, explain, and execute scripts: ``` You: "I need a script that backs up my Documents folder to an external drive every day" Machine: "I'll create a backup script for you. Here's what it will do: 1. Check if your external drive is connected 2. Create a timestamped backup folder 3. Use rsync to copy only changed files 4. Log the backup status Would you like me to create this script now?" ``` ### Automation Workflows Create complex automation workflows through conversation: ``` You: "I want to set up a workflow that downloads data from our website analytics, processes it with Python, and emails me a weekly report" Machine: "I'll help you create that automation. Let me break this down into steps: 1. Use curl to download analytics data 2. Process it with a Python script 3. Create a report document 4. Send it via email 5. Schedule this with cron/Task Scheduler Would you like me to help you implement each step?" ``` ## Security Considerations When using command-line functionality, Machine follows these security principles: ### Permission Requirements * Machine only executes commands with your current user permissions * Elevated privileges (sudo/Administrator) are clearly indicated before execution * You must explicitly approve commands requiring elevated privileges ### Command Review * All commands are shown to you before execution * Potentially destructive operations are highlighted with warnings * Machine explains complex commands in plain language ### Execution Contexts * By default, commands run in a secure execution context * File system access is limited to your user directories unless explicitly expanded * Network operations follow strict security policies ## Best Practices Follow these guidelines for optimal and secure use of command-line functionality: 1. **Review commands carefully** before approving execution 2. **Start simple** with basic commands before attempting complex operations 3. **Use descriptive requests** to help Machine understand your goals 4. **Ask for explanations** if you don't understand what a command does 5. **Create scripts** for recurring operations rather than executing complex commands repeatedly ## Platform-Specific Considerations ### Windows * PowerShell and Command Prompt commands are both supported * Some operations may require Administrator privileges * UNC paths are supported for network operations ### macOS/Linux * Bash, Zsh, and other common shells are supported * Environment variables are preserved between commands * File permissions follow standard Unix models For advanced command-line operations or custom scripts, you can provide Machine with specific syntax or ask it to optimize commands for your particular environment. # Frequently Asked Questions Source: https://support.myapps.ai/machine/faq Common questions about Machine # Frequently Asked Questions Find answers to common questions about Machine, its capabilities, and how to get the most out of your experience. ## General Questions Machine is an AI assistant that helps you accomplish real-world tasks through natural conversation. It combines powerful capabilities like browser automation, file management, command-line execution, and API integration to solve complex problems and automate workflows. Unlike many AI assistants that can only provide information or generate content, Machine can actively perform tasks on your behalf. It can browse the web, create and edit files, run system commands, and integrate with external services—all through simple conversations. No! Machine is designed to be accessible to users of all technical skill levels. You can communicate in plain language, and Machine will handle the technical details. Of course, users with technical knowledge can provide more specific instructions when needed. Yes, Machine works on smartphones and tablets through our responsive web interface. We also offer dedicated mobile apps for iOS and Android with optimized experiences for on-the-go use. ## Account & Access No! Machine is designed to be accessible to users of all technical skill levels. You can communicate in plain language, and Machine will handle the technical details. Of course, users with technical knowledge can provide more specific instructions when needed. Yes, Machine works on desktop computers, laptops, tablets, and smartphones. Your conversations and files sync automatically across all your devices, so you can seamlessly switch between them. Usage limits depend on your subscription plan. The Basic plan includes specific monthly limits on conversations and tasks, while Professional and Enterprise plans offer higher or unlimited usage. You can view your current usage in the Account Settings. To create an account, visit [machine.myapps.ai](https://machine.myapps.ai) and click "Sign Up." You can register using your email address, Google account, or Microsoft account. Follow the prompts to complete your profile setup. Yes, you can upgrade or downgrade your subscription at any time. Go to Account Settings → Subscription to view available plans and make changes. When upgrading, the new features become available immediately. When downgrading, the change takes effect at the end of your current billing cycle. If you've forgotten your password, click "Forgot Password" on the login screen. Enter your email address, and we'll send you a secure link to reset your password. For security reasons, the link expires after 30 minutes. Each Machine account is designed for individual use. One account corresponds to one user. For different users, separate Machine accounts are required. ## Billing & Usage Machine offers monthly and annual subscription plans. Monthly plans are billed every 30 days, while annual plans are billed once per year with a discount. Your subscription automatically renews until canceled. We accept major credit cards (Visa, Mastercard, American Express, Discover), PayPal, and Apple Pay. Enterprise customers can also pay by invoice with net-30 terms. Each subscription tier includes a set amount of usage across different features. For example, the Standard plan includes 100 hours of automation per month. Usage resets on your billing date. If you need more usage, you can upgrade your plan or purchase additional usage as needed. Yes, you can cancel your subscription at any time from Account Settings → Subscription → Cancel Subscription. When you cancel, you'll continue to have access to your current plan until the end of your billing period. We offer a 7-day money-back guarantee for new subscriptions. If you're not satisfied with Machine within the first week, contact support for a full refund. After this period, refunds are handled on a case-by-case basis. ## Files & Data Machine supports most common file formats, including: * Documents: PDF, DOCX, TXT, RTF, HTML, MD * Spreadsheets: XLSX, CSV, TSV * Images: JPG, PNG, GIF, BMP, SVG * Code: Most programming and markup languages * Audio: MP3, WAV, OGG, FLAC * Video: MP4, MOV, AVI (limited analysis capabilities) If you need to work with a specific file type not listed here, you can ask Machine if it's supported. Each subscription tier includes different storage limits. The maximum file size for uploads is 100MB per file. Enterprise plans offer expanded storage options for greater needs. Machine can only access files you explicitly upload or create within the platform. It cannot directly access your local file system without your permission, and all file operations are securely handled within your workspace. ## Technical Requirements Machine works on all modern browsers, including: * Google Chrome (recommended) * Mozilla Firefox * Microsoft Edge * Safari * Opera We recommend keeping your browser updated to the latest version for optimal performance. Machine's web interface has minimal system requirements: * 4GB RAM (8GB recommended) * Modern web browser * Stable internet connection (minimum 5Mbps) For demanding automation tasks, a more powerful system may improve performance. Machine requires an internet connection to function as it runs in the cloud. However, some capabilities are cached for limited offline use, and you can access previously created files while offline. ## Features & Capabilities Yes! Machine can generate code in various programming languages based on your requirements. It can create complete applications, scripts, or code snippets, and can help debug existing code. Machine's interface runs in your browser, but the core technology operates on secure cloud servers. This architecture provides powerful capabilities without taxing your local device's resources. Yes, with Professional and Enterprise plans, you can schedule automation tasks to run on a recurring basis—hourly, daily, weekly, or monthly—without manual intervention. ## Security & Privacy Yes, security is our top priority. All data is encrypted in transit and at rest. We implement strict access controls, regular security audits, and follow industry best practices. For more details, see our [Security & Privacy](/machine/security-privacy) page. Conversations are stored to maintain context and improve your experience. You can control data retention in your privacy settings, and you have the option to disable conversation history if preferred. Yes, you can delete specific conversations, files, or your entire account data at any time through your account settings. When you delete your data, it's permanently removed from our systems according to our data retention policy. ## Troubleshooting If Machine is running slowly: 1. Refresh your browser 2. Clear your browser cache 3. Check your internet connection 4. Try a different browser 5. Reduce the number of open workspaces If problems persist, contact our support. Common errors and their solutions are documented in our [Machine Errors](/machine/machine-errors) page. If you encounter an error not covered there, please contact support with the specific error message and what you were trying to do. We welcome your feedback! To report a bug or suggest a feature: 1. Click on the "Feedback" button in the bottom right corner 2. Select "Report a Bug" or "Suggest a Feature" 3. Fill out the form with as much detail as possible We review all submissions and will follow up if we need additional information. ## Still Have Questions? If your question isn't answered here, please contact our support: * **Email**: [rcohen@mytsi.org](mailto:rcohen@mytsi.org) * **Chat**: Click the "Help" button in the Machine interface * **Knowledge Base**: Explore more detailed articles at [help.machine-ai.com](https://help.machine-ai.com) Our support is available Monday through Friday, 9 AM to 6 PM Eastern Time, with limited weekend coverage for urgent issues. # Machine Features Source: https://support.myapps.ai/machine/features Comprehensive overview of Machine capabilities # Machine Features & Capabilities Machine is an AI assistant that helps you accomplish real-world tasks through natural conversation. This page provides an overview of Machine's powerful capabilities that enable you to solve complex problems with ease. ## Core Capabilities Interact with Machine through simple, natural dialogue—just like chatting with a colleague. Let Machine handle repetitive tasks while you focus on more important work. From data analysis to research, Machine helps solve complex problems step-by-step. Machine evolves with each interaction, becoming more attuned to your specific needs. ## Feature Suite ### Browser Automation Machine can navigate the web on your behalf, extracting relevant information and performing tasks across websites. * **Web Navigation**: Machine can browse websites, click links, fill forms, and extract data * **Automated Research**: Gather information from multiple sources and synthesize findings * **Multistep Workflows**: Execute complex sequences of actions across different websites ### File Management Create, edit, and organize documents directly through conversation. * **Document Creation**: Generate various file types including text documents, spreadsheets, and presentations * **File Editing**: Update existing documents with new content or formatting * **File Organization**: Rename, move, and organize files based on your specifications ### Web Crawling & Search Powerful search capabilities that extend beyond standard search engines. * **Deep Web Search**: Access and extract information from sources not easily accessible through standard search * **Comprehensive Research**: Gather information from multiple sources for thorough analyses * **Content Monitoring**: Track changes to websites and receive notifications about updates ### Command-Line Execution Execute system-level tasks and commands through simple conversations. * **System Operations**: Run commands to automate local system tasks * **Script Execution**: Create and run scripts to perform complex operations * **System Monitoring**: Check system status and receive notifications about important events ### Website Deployment Build and deploy websites through conversational guidance. * **Site Creation**: Generate website code based on your specifications * **Deployment Assistance**: Navigate hosting platforms and deploy your site * **Site Maintenance**: Make updates and monitor your website's performance ### API Integration Connect with external services and data sources through various APIs. * **Data Exchange**: Pull from and push data to external systems * **Service Integration**: Connect with third-party tools and platforms * **Workflow Automation**: Create automated processes that span multiple services ## Getting Started To begin using Machine's powerful features, simply start a conversation about what you're trying to accomplish. Machine will guide you through the process, asking clarifying questions when needed and providing updates as tasks progress. Machine respects privacy and security protocols. It will always ask for explicit permission before accessing sensitive information or performing actions that impact your systems or data. # File Management Source: https://support.myapps.ai/machine/file-management Working with files and documents through Machine # File Management Machine provides powerful file management capabilities that enable you to create, edit, organize, and process files through natural conversation. This eliminates the need to switch between multiple applications when working with documents. ## Supported File Types Machine can work with a wide variety of file types across different categories: ### Documents .txt, .md, .rtf, .log .docx, .xlsx, .pptx, .pdf .py, .js, .html, .css, .json ### Media .jpg, .png, .gif, .svg .mp3, .wav, .m4a, .ogg .mp4, .mov, .avi, .mkv ### Data .csv, .tsv, .xlsx .sql, .db, .sqlite .json, .xml, .yaml ## File Operations ### Creating Files Ask Machine to create new files by describing what you need: ``` You: "I need a spreadsheet to track my monthly expenses" Machine: "I can create an expense tracking spreadsheet for you. Would you like me to include categories like housing, transportation, food, etc.?" ``` ``` You: "Create a Python script that sorts files in a directory by type" Machine: "I'll create a Python script that organizes files by their extensions. Do you want it to move the files into separate folders or just display the sorted list?" ``` ### Editing Files Machine can help you edit existing files: ``` You: "Update my resume to include my new position at Acme Corp" Machine: "I'll help you update your resume. Could you upload the current version so I can make the changes?" ``` ``` You: "Add a new section to my project documentation about the API endpoints" Machine: "I'll add an API endpoints section to your documentation. What specific endpoints should I include, and do you want examples for each one?" ``` ### File Conversion Convert between different file formats: ``` You: "Convert this CSV file to JSON" Machine: "I'll convert your CSV data to JSON format. Would you like me to create a nested structure or keep it as a flat array of objects?" ``` ``` You: "Turn this markdown document into a PDF with proper formatting" Machine: "I'll convert your markdown to a nicely formatted PDF. Would you like me to include a table of contents and apply any specific styling?" ``` ### File Analysis Extract insights and information from your files: ``` You: "Summarize the key points from this research paper" Machine: "I'll analyze the research paper and extract the main findings, methodology, and conclusions. Would you like me to focus on any specific aspects?" ``` ``` You: "Analyze this dataset and tell me what patterns you notice" Machine: "I'll analyze your dataset for patterns and trends. I can create visualizations, calculate statistics, and identify correlations. Any specific insights you're looking for?" ``` ## Workspace Management ### Organizing Files Machine helps you keep your files organized: ``` You: "Help me organize my project files into a logical folder structure" Machine: "I'll help organize your project files. Based on what I see, I recommend creating folders for documentation, source code, resources, and tests. Would that work for you?" ``` ### Batch Operations Perform operations across multiple files: ``` You: "Rename all the image files in this folder to follow the pattern project-image-001.jpg" Machine: "I'll rename your image files with the pattern 'project-image-001.jpg', 'project-image-002.jpg', etc. There are 24 images. Should I proceed?" ``` ### **Downlaoding FIles** Download multiple files: image.png ## File Organization Tips * Create a consistent folder structure for your projects * Use descriptive file names that include dates where appropriate * Keep related files together in project-specific folders * Use tagging and labels for easier search and retrieval * Regularly review and archive older files ## File Sharing ### Sharing Options Machine provides several ways to share your files: * **Direct download links**: Share files with anyone via secure links * **Export to cloud services**: Send files to Google Drive, Dropbox, or other services ``` You: "Share this presentation" Machine: "How would you like to share the presentation? I can create a direct download link or export it to a cloud service like Google Drive." ``` If you have specific file management requirements, you can ask Machine to recommend an organization system tailored to your needs and workflow. ## Best Practices For optimal file management with Machine: 1. **Be specific with filenames**: Use descriptive names for easier searching and organization 2. **Leverage metadata**: Add tags and descriptions to important files 3. **Use workspaces**: Organize related files into dedicated workspaces for improved context 4. **Regular cleanup**: Schedule periodic reviews of your files to archive or delete unnecessary items 5. **Template creation**: Create templates for frequently used file types ## File Security Machine protects your files with multiple security measures: * **Encryption**: All files are encrypted in transit and at rest * **Access controls**: Fine-grained permissions for your files * **Virus scanning**: Automatic scanning of all uploaded files * **Version history**: Recover previous versions of your files if needed File storage quotas depend on your subscription plan. See the [Pricing & Plans](/machine/pricing-plans) page for details on storage limits for each tier. # Getting Started Source: https://support.myapps.ai/machine/getting-started Your first steps with Machine # Getting Started with Machine Welcome to Machine! This guide will help you start your journey with our AI assistant designed to help you accomplish real-world tasks through natural conversation. ## First Interaction When you first launch Machine, you'll be greeted with a simple conversation interface. There's no complex setup required—just start typing as if you were talking to a helpful colleague. Visit [machine.myapps.ai](https://machine.myapps.ai) and log in with your credentials. If you don't have an account yet, you can create one in minutes. Begin by describing what you're trying to accomplish. For example: * "I need to research the latest trends in renewable energy" * "Help me analyze this customer feedback data" * "I need to build a simple website for my portfolio" Machine will ask clarifying questions and guide you through the process. Answer its questions to help it better understand your needs. Before taking significant actions, Machine will ask for your approval. Review its proposals carefully before proceeding. ## Understanding Machine's Capabilities The best way to learn what Machine can do is to simply ask it. For example: * "What can you help me with?" * "Can you help me build a website?" * "How would you approach analyzing this dataset?" Machine will explain its capabilities relevant to your task and suggest a path forward. ## File Uploading Many tasks require working with your files. You can easily upload files for Machine to work with. 1. Click the paperclip icon in the chat interface 2. Select the file(s) you want to upload 3. Wait for the upload to complete 4. Machine will acknowledge the file and ask what you'd like to do with it Machine can work with a wide variety of file types, including: * Documents (.docx, .pdf, .txt, etc.) * Spreadsheets (.xlsx, .csv, etc.) * Images (.jpg, .png, .svg, etc.) * Data files (.json, .xml, etc.) * And many more ## Workspace Management Your Machine workspace helps you organize your projects and files for easy access. ### Creating a New Workspace 1. Click on the "Workspaces" option in the sidebar 2. Select "New Workspace" 3. Give your workspace a name and description 4. Start adding files and conversations to your workspace ### Managing Files in Your Workspace * All files you upload are automatically saved to your current workspace * You can organize files into folders by asking Machine to do so * Access your files anytime by clicking the "Files" tab in your workspace ## Best Practices To get the most out of Machine, follow these best practices: 1. **Be specific**: The more details you provide, the better Machine can help you. 2. **Start simple**: Begin with straightforward tasks and gradually move to more complex ones as you become familiar with Machine's capabilities. 3. **Provide feedback**: Let Machine know if its approach isn't working or if you need to adjust course. 4. **Use projects**: Organize related tasks into projects to maintain context across multiple sessions. 5. **Ask for explanations**: If you're not sure why Machine is taking a particular approach, just ask for clarification. ## Next Steps Once you're comfortable with the basics, explore these resources to take your Machine experience to the next level: Discover the full range of Machine's capabilities. Get inspired by examples of how others are using Machine. Learn how to connect Machine to external services and APIs. Find solutions to common issues. Remember, Machine learns from your interactions. The more you use it, the better it gets at understanding your preferences and requirements. # Machine Errors Source: https://support.myapps.ai/machine/machine-errors Common error codes and solutions # Error Reference Guide This document provides a list of common errors that users might encounter when using Machine, along with their causes and solutions. ## Common Error Codes **Resource Not Found** The requested resource could not be found or has been removed. **Solution**: * Check that you're using the correct URL * Try refreshing the page * Return to the dashboard and try accessing the resource again **Request Error** Machine couldn't process your request due to validation issues. **Solution**: * Try simplifying your request * Check that any uploaded files meet our requirements * If asking a complex question, break it into smaller parts ## Account & Subscription Errors ### "Invalid subscription tier" Error **Cause**: Your subscription appears to be invalid or has encountered a processing issue. **Solution**: 1. Go to Account Settings → Subscription to verify your current plan 2. Try refreshing your browser or logging out and back in 3. Contact customer support if the issue persists ### Checkout Problems **Cause**: Issues during the payment or subscription upgrade process. **Solution**: 1. Try a different browser or clear your browser cache 2. Ensure your payment method is valid and has sufficient funds 3. Use our direct checkout link from your account page ## Usage Problems ### "Session Expired" Error **Cause**: Your session has timed out due to inactivity or system maintenance. **Solution**: 1. Refresh the page 2. Log out and log back in 3. If the problem continues, wait a few minutes and try again ### "File Upload Failed" Error **Cause**: Issues with file uploads can occur for several reasons. **Solution**: 1. Ensure your file is under the 25MB size limit 2. Check that you're using a supported file format 3. Try uploading again or use a different file ### "Chat Not Responding" Error **Cause**: The AI assistant might be processing a complex request or experiencing high demand. **Solution**: 1. Wait a few moments for the response 2. If no response after 30 seconds, refresh the page 3. Start a new chat if the issue persists ## Feature Access Issues ### "Feature Not Available" Error **Cause**: You've attempted to use a feature not included in your current subscription plan. **Solution**: 1. Check your current subscription level in Account Settings 2. Upgrade your plan to access premium features 3. See our pricing page for feature comparison ### "Usage Limit Reached" Error **Cause**: You've reached the maximum usage for your current subscription period. **Solution**: 1. Check your usage statistics in Account Settings 2. Wait until your usage resets on your billing date 3. Consider upgrading to a plan with higher limits ### "Tool Execution Timeout" Error **Cause**: You may encounter errors like `ToolResult(success=False, output="Error executing command: Failed to execute session command: HTTPSConnectionPool(host='app.daytona.io', port=443): Read timed out. (read timeout=179.9997366879834)")` when Machine attempts to run commands or code. **Solution**: 1. This error is normal and simply means the AI failed or code it wrote failed to execute within the expected time 2. Just ask Machine to retry the operation 3. If the error persists, try breaking down your request into smaller steps For any errors not covered in this guide, please contact our customer support at [rcohen@mytsi.org](mailto:rcohen@mytsi.org) or use the in-app chat support for assistance. # OpenClaw Gateway Source: https://support.myapps.ai/machine/open-claw Connect and manage messaging channels with your Machine agents via OpenClaw # OpenClaw Gateway Connect your Machine agents to messaging platforms — Telegram, Discord, WhatsApp — through a single gateway. OpenClaw gives you one place to select an agent, configure channels, and control your gateway. ## Overview Choose which agent to configure for messaging channels. That agent's instructions and model power all connected channels. Connect and manage Telegram, Discord, and WhatsApp from one dashboard. Gateway shows **Active** when ready. When the gateway is active, use **Open Gateway UI** to manage your OpenClaw instance and see real-time status. ## Machine Backend Configuration Before connecting channels, configure how OpenClaw talks to Machine. Set the **default model** used by this gateway. Changing it will restart the gateway so the new model takes effect. Only required for certain models. Enter your key in the format `sk-...`. Need a key? Get one from your provider. Click **Save Machine Config** after updating. The gateway will use these settings for all messaging channels. ## Connecting Messaging Channels ### Telegram Create a bot if you don't have one yet. Follow BotFather's instructions to name your bot and choose a username (must end in `bot` or `Bot`). BotFather provides a token like `1234567890:ABCdefGHIjklMNOpqrsTUVwxyz`. Copy it fully. In the Telegram section, paste the token and click **Connect Telegram Bot**. Start a chat with your bot in Telegram; you'll receive a **pairing code**. Enter it in "Enter pairing code from Telegram" and click **Approve Pairing**. After starting a chat with your bot, you'll receive a pairing code. Enter it in the gateway UI to approve the connection. ### Discord & WhatsApp Use the same OpenClaw Gateway UI to add Discord and WhatsApp. Each channel has its own connect flow; follow the on-screen steps for each platform. ## Gateway Control Manage your OpenClaw instance from the gateway dashboard. Open the full OpenClaw gateway interface in a new tab. Quick link to OpenClaw Studio for advanced configuration. Access the gateway console for logs and debugging. Check for and apply OpenClaw updates. ### Interactive Terminal (Recommended) **Open Machine Terminal** gives full support — works with `openclaw onboard`, `openclaw config --section channels`, and all interactive commands. No timeout. Use the terminal for: * `openclaw onboard` — guided setup * `openclaw config --section channels` — channel configuration * Any other interactive OpenClaw commands ### Gateway Actions Generate a new gateway URL if the current one has expired or changed. Resolve origin/CORS issues if the gateway fails to load in the browser. Restart the OpenClaw gateway process. Use after config changes or if the gateway is unresponsive. Update the displayed gateway status (e.g. Running / Stopped). Enable to automatically approve pairing requests for connected channels (e.g. Telegram). Start the gateway if it is stopped. Required before messaging channels can receive messages. ## Gateway URL & Status After the gateway is running, you'll see: * **Gateway URL** — A URL like `https://...proxy.daytona.works?token=...`. This is how messaging platforms reach your OpenClaw instance. Keep it private. * **Status** — Typically **Running** when the gateway is active. Do not share your gateway URL or token publicly. Anyone with the URL could send traffic to your gateway. ## Timeout & Sandbox The gateway auto-stops after a period of inactivity (e.g. **1 hour**) to save resources. The current setting is shown on the dashboard. To change the timeout, you may need to delete and recreate the setup. If the gateway has stopped due to inactivity, click **Wake Sandbox** to reactivate it without recreating the setup. ## Help & Troubleshooting Use the **Help & Troubleshooting** section in the OpenClaw Gateway UI for: * Machine backend configuration issues * Channel connection and pairing problems * Gateway URL and origin errors * OpenClaw CLI and terminal usage ## Removing the Setup Click **Delete Messaging Setup** to remove the OpenClaw gateway and all channel connections. You can create a new setup later; you'll need to reconnect each channel (Telegram, Discord, WhatsApp) and re-enter tokens or pairing codes as required. Deleting the setup does not delete your agents — only the gateway and its channel links are removed. ## Summary | Action | Where | | ------------------------------- | ---------------------------------------------------------------------------------------- | | Select agent for messaging | OpenClaw Gateway → **Select Agent** | | Configure Machine model/API key | **Machine Backend Configuration** → **Save Machine Config** | | Connect Telegram | Paste bot token → **Connect Telegram Bot** → **Approve Pairing** with code from Telegram | | Open full gateway UI | **Open Gateway UI** | | Use CLI (no timeout) | **Open Machine Terminal** | | Reactivate after auto-stop | **Wake Sandbox** | | Remove gateway and channels | **Delete Messaging Setup** | # Pricing & Plans Source: https://support.myapps.ai/machine/pricing-plans Subscription options for Machine # Pricing & Plans Machine offers flexible subscription plans designed to fit your needs, whether you're an individual user or an organization with advanced requirements. ## Plan Comparison **\$35/month** Perfect for individual users getting started with AI automation * 500 hours of usage per month * 10GB storage * Standard response speed * Core features access * Email support **\$200/month** Ideal for professionals who need more power and flexibility * 3000 hours of usage per month * 50GB storage * Priority response speed * All features including scheduling * Priority email support * Advanced API access **Custom pricing** Tailored solutions for organizations with specialized needs * Unlimited usage * Custom storage limits * Fastest response speed * All features plus exclusive enterprise tools * Dedicated support manager * SSO and advanced security * Custom integrations Save 20% with annual billing on all plans! [Contact us](mailto:rcohen@mytsi.org) for special pricing for educational institutions and non-profits. ## Feature Availability | Feature | Basic | Professional | Enterprise | | ------------------------ | ------- | ------------ | ------------- | | Natural conversation | ✓ | ✓ | ✓ | | File creation & editing | ✓ | ✓ | ✓ | | Web search | ✓ | ✓ | ✓ | | Basic browser automation | ✓ | ✓ | ✓ | | Command-line operations | Limited | Full | Full + Custom | | Basic API integration | ✓ | ✓ | ✓ | | Feature | Basic | Professional | Enterprise | | --------------------------- | ------- | ------------ | ------------- | | Advanced browser automation | - | ✓ | ✓ | | Task scheduling | - | ✓ | ✓ | | Custom workflows | - | ✓ | ✓ | | Multi-step automations | - | ✓ | ✓ | | Advanced API integration | - | ✓ | ✓ | | Custom code execution | Limited | Full | Full + Custom | | Feature | Basic | Professional | Enterprise | | ------------------------- | ----- | ------------ | ---------- | | Data encryption | ✓ | ✓ | ✓ | | Custom retention policies | - | ✓ | ✓ | | GDPR compliance | ✓ | ✓ | ✓ | | HIPAA compliance | - | - | ✓ | | SSO integration | - | - | ✓ | | VPC deployment | - | - | ✓ | | Feature | Basic | Professional | Enterprise | | ----------------- | ----- | ------------ | ---------- | | Email support | ✓ | ✓ | ✓ | | Priority queue | - | ✓ | ✓ | | Phone support | - | - | ✓ | | Dedicated manager | - | - | ✓ | | Training sessions | - | 1 per year | 4 per year | | SLA guarantees | - | - | ✓ | ## Usage Details ### Hours of Usage Usage is measured in active interaction hours: * **Active interaction time**: When Machine is actively working on tasks for you * **Idle time**: Not counted against your usage limit * **Background tasks**: Count as active time even when you're not directly interacting Usage does not roll over from month to month. Your usage resets on your billing date. ### Storage Capacity Storage limits apply to all files created or uploaded within Machine: * **File uploads**: Count against your storage limit * **Generated content**: Files created by Machine count against your limit * **Archives**: You can archive files to free up active storage space ## Upgrading and Downgrading You can change your plan at any time from your account settings: * **Upgrades**: Take effect immediately, with prorated billing for the remainder of your current billing cycle * **Downgrades**: Take effect at the end of your current billing cycle * **Enterprise transitions**: Require coordination with our department ## Additional Resources Learn about our custom enterprise offerings, including on-premises deployment, custom integrations, and dedicated support. Special pricing available for educational institutions, students, and faculty members. Qualifying non-profit organizations can access discounted or free plans to support their mission. Join our partner program to offer Machine to your clients or integrate with your existing solutions. ## Frequently Asked Questions Start with the plan that matches your expected usage. The Basic plan is perfect for individuals exploring Machine's capabilities. Professional is ideal for regular users who need more power and features. Enterprise is designed for organizations with specialized needs. If you approach your usage limit, you'll receive notifications at 80% and 95% of your allocation. If you exceed your limit, you have two options: 1. Purchase additional usage for the current month 2. Upgrade to a higher tier plan We'll never automatically charge you for overages without your approval. Yes! We offer a 7-day free trial of the Professional plan for all new users. No credit card required to start your trial. You can also explore our interactive demo without creating an account to see Machine in action. We currently bill in USD, but your payment method will handle the currency conversion based on current exchange rates. Enterprise customers can arrange for billing in select additional currencies. All prices listed are in USD. Prices may vary by region due to local taxes and regulations. For the most current pricing information, please refer to your account page. # Security & Privacy Source: https://support.myapps.ai/machine/security-privacy How Machine protects your data and information # Security & Privacy Machine is built with a security-first approach, ensuring your data, conversations, and activities remain protected at all times. ## Data Protection ### Encryption All communication between your devices and Machine uses TLS 1.3 encryption, ensuring that your data is secure as it travels across the internet. Your stored data is encrypted using AES-256, protecting it from unauthorized access even in the unlikely event of a breach. Sensitive operations and files can be end-to-end encrypted, meaning only you can access the unencrypted data. We use industry-standard key management services to securely store and rotate encryption keys. ### Access Controls * **Role-Based Access**: Only authorized personnel have access to systems containing user data * **Least Privilege Principle**: Personnel are granted only the minimum permissions necessary * **Multi-Factor Authentication**: Required for all internal systems * **Access Logging**: All access to user data is logged and audited ## Privacy Guarantees ### Data Usage Machine only uses your data for the following purposes: 1. **Providing our services**: Processing your requests and performing tasks you explicitly ask for 2. **Improving our services**: Learning from interactions to improve response quality 3. **Security and fraud prevention**: Detecting and preventing fraudulent activity and security threats ### Data Sharing We do not sell or rent your personal information. We only share your information in the following limited circumstances: * With your explicit consent * To comply with legal obligations * With service providers who help us deliver our services (under strict confidentiality agreements) ### Data Retention * **Active Conversations**: Retained to provide context for ongoing interactions * **Completed Tasks**: Retained for 30 days by default, or based on your configured retention policy * **Account Information**: Maintained as long as you have an active account * **Data Deletion**: You can request deletion of your data at any time through your account settings ## Security Measures ### Infrastructure Security * **Cloud Security**: We leverage enterprise-grade cloud infrastructure with built-in security features * **Network Isolation**: Critical systems are isolated from public networks * **DDoS Protection**: Advanced systems to prevent denial-of-service attacks * **Vulnerability Management**: Regular scanning and timely patching of vulnerabilities ### Application Security * **Secure Development**: We follow secure coding practices throughout our development lifecycle * **Regular Audits**: Our codebase undergoes regular security audits and penetration testing * **Dependency Scanning**: We scan for and update vulnerable dependencies * **Bug Bounty Program**: We work with security researchers to identify and fix security issues ### Operational Security * **Security Monitoring**: 24/7 monitoring for suspicious activities * **Incident Response**: Defined procedures to quickly address security incidents * **Regular Drills**: Our team conducts regular security incident response drills * **Vendor Assessment**: All third-party vendors undergo security assessment ## Compliance Machine is designed to help you maintain compliance with various regulations and standards: * **GDPR**: Compliant with the European Union's General Data Protection Regulation * **CCPA**: Compliant with the California Consumer Privacy Act * **SOC 2**: We maintain SOC 2 Type II certification * **HIPAA**: Available on Enterprise plans for customers who require HIPAA compliance ## Reporting Security Issues If you discover a security vulnerability, please report it to [rcohen@mytsi.org](mailto:rcohen@mytsi.org). We appreciate your help in keeping Machine secure and will respond promptly to your report. ## Privacy Controls Machine gives you control over your privacy: Access your privacy settings by clicking on your profile > Settings > Privacy. From there, you can: * Adjust data retention periods * Enable/disable usage tracking * Manage conversation history * Control data sharing preferences You can export all your data at any time: 1. Go to Settings > Privacy > Export Data 2. Select what data you want to export 3. Choose your preferred format (JSON, CSV, HTML) 4. Submit your request You'll receive a download link when your export is ready. To delete your data: 1. Go to Settings > Privacy > Delete Data 2. Select what data you want to delete 3. Confirm your decision Please note that deletion is permanent and cannot be undone. For more detailed information about our privacy practices, please refer to our [Privacy Policy](https://machine.myapps.ai/privacy). # Telegram Integration Source: https://support.myapps.ai/machine/telegram Connect your Machine agents to Telegram and chat with them from anywhere # Telegram Setup Guide Connect your Machine agents to Telegram and chat with them from anywhere — in about 5 minutes. ## How It Works Machine's Telegram Integration lets you connect any of your Machine agents to a Telegram bot. Once connected, anyone who messages your bot on Telegram will receive responses powered by your agent — complete with the model and instructions you've configured. User sends a message to your Telegram bot Telegram forwards the message to Machine via webhook Machine routes the message to your selected agent Your agent processes the message using its configured model and instructions The response is sent back to the user on Telegram ## Prerequisites A personal Telegram account (free) An active Machine account with at least one agent At least one agent configured (powers your bot's responses) ## Part 1 — Create a Bot on Telegram \~2 min BotFather is Telegram's official tool for creating and managing bots. Open Telegram and search for **@BotFather** (look for the blue verified checkmark ✓). Tap **Start** or type `/start`. BotFather will reply with a list of available commands. Send the command `/newbot`. BotFather asks: "How are we going to call it?" — type any display name you like, e.g. **MyAssistantBot**. The username must end in **bot** or **Bot**. If it's taken, BotFather will ask you to try another. BotFather responds with "Done! Congratulations on your new bot." and provides: * A link to talk to your bot — e.g. `t.me/MyAssistantBot` * An API token — looks like `123456789:ABCdefGHIjklMNO...` Keep your token secret! Anyone with this token can control your bot. If it ever gets leaked, revoke it with `/revoke` in BotFather. ## Part 2 — Connect the Bot in Machine \~1 min Now bring that bot token into Machine to link it with one of your agents. In the Machine dashboard sidebar, click **Telegram** — or go directly to `/telegram`. A configuration dialog will open with three fields. In the Bot Token field, paste the API token from BotFather. Format: `123456789:ABCdefGHIjklMNO...` Choose which Machine agent should handle incoming messages. This agent's instructions, personality, and tools will power your bot's responses. Optionally pick a specific AI model. If you leave this blank the bot uses the agent's default model. Machine validates your token, sets up the webhook automatically, and shows a success notification. Your bot appears in the Connected Bots table with a green **Active** badge. ## Part 3 — Talk to Your Bot Your bot is now live! Open Telegram, search for your bot's username (e.g. **@MyAssistantBot**), and tap **Start**. ``` @MyAssistantBot online /start 👋 Hello! I'm your Machine-powered assistant. Here's what I can help with… Hey, you're my Telegram bot That's right! I'm connected to your Machine agent. How can I help you today? What can you do? I can help with a variety of tasks — just ask me anything and I'll use all the tools and knowledge my agent has access to! ``` The bot maintains conversation context within each chat thread, so you can have natural back-and-forth conversations just like chatting with a person. ## Managing Your Bots All your connected bots appear on the Telegram dashboard. Click the pencil icon next to a bot, pick a new model, and save. The change takes effect immediately for new messages. Click the trash icon and confirm. The bot stops responding, but you can reconnect it any time with the same token. Machine supports multiple bots per account. Each can have its own agent and model — just click "Add Bot" again. ## Tips & Best Practices ### Agent Configuration Customize your agent's instructions for Telegram's conversational style. Add instructions like "Keep responses concise and suitable for mobile messaging" Telegram messages tend to be shorter — tune your prompts accordingly ### Security Never share your bot token publicly — treat it like a password. If compromised: @BotFather → `/revoke`. Machine encrypts your bot token at rest. ### BotFather Commands Change display name Set bot description Upload profile pic Define slash commands Toggle privacy mode Regenerate token ## Advanced Branding & Groups ### Branding Your Bot Use these BotFather commands to give your bot a professional identity: **Profile Picture** - Upload a 512x512 square logo **Description** - Shown before users start the bot **About Text** - Shown on the bot's profile page **Menu List** - Define your bot's slash command menu ### Group Chat Intelligence When you add your bot to a group, Machine provides it with context about who is speaking. By default, bots only see messages that @mention them. To see all messages, use `/setprivacy` → Disable. Note: Changing privacy requires removing and re-adding the bot to the group. Disabling privacy increases credit usage as the bot processes all messages in the group. ## Troubleshooting The webhook may not have been set correctly. Try disconnecting and re-adding the bot. Make sure you copied the full token from BotFather. It should look like `123456789:ABCdefGHIjklMNO…` Ensure an agent is selected in the Machine dashboard. If you revoked the token in BotFather, you'll need a new one. Response time depends on the chosen model and agent complexity. Some models are faster than others. Privacy Mode is on by default — use `/setprivacy` in BotFather to disable it, then remove and re-add the bot to the group. If your bot stops responding, another service might have claimed the webhook. Simply click "Save Configuration" again in Machine to re-claim it. ## Technical Specifications ### Message Delivery Strict **4,096 character limit** per message. Machine automatically truncates longer responses. Formatting is intentionally simplified to prevent Telegram parse errors. Bold/Italic are converted to plain text. ### Vision & Media Supported if using a vision-capable model. Send photos directly or as uncompressed files. Currently supports image-based documents. Non-image file processing is coming soon. ## Frequently Asked Questions Yes! Machine supports multiple bots per account. Each can have its own agent and model. Yes, you can use the same agent for multiple Telegram bots. Yes, the bot maintains conversation context within each chat thread. Click the pencil icon next to your bot in the Telegram dashboard, select a different agent, and save. Messages consume credits based on your Machine plan and the model used by your agent. Currently, bots work best in direct messages and groups. Channel support may vary. If you delete the bot in BotFather, you'll need to create a new bot and reconnect it in Machine. ## Ready to Get Started? Head to the Telegram dashboard to connect your first bot in minutes. Connect your first bot in Machine Create your Telegram bot # Machine 101 Source: https://support.myapps.ai/machine/the-machine The Machine # About Machine Machine is the world's first fully autonomous AI agent capable of performing complex tasks without human supervision. Founded in 2025, we're on a mission to create AI systems that can work independently to solve problems across domains ranging from data analysis to coding, research to content creation. ## Key Features Machine handles complex tasks on your behalf, learning your preferences and adapting to your needs over time. Perfect for individuals starting with autonomous AI ## Explore our plans Perfect for individuals and small teams starting with autonomous AI Advanced features for businesses requiring more sophisticated automation Fully customizable solutions for organizations with complex needs ## Get Started Ready to experience the power of fully autonomous AI? [Contact us](mailto:rcohen@mytsi.org) to schedule a demo or learn more about how Machine can transform your workflows. # Machine Tools Source: https://support.myapps.ai/machine/tools Official tools available in Machine and how to use them # Machine Tools Machine provides a powerful set of built-in tools that can be called directly in conversations. These tools enable you to perform complex operations like file management, browser automation, deployments, and more. ## File Management Tools ### execute-command Execute terminal or shell commands directly from Machine. ``` You: "Can you run 'npm install' in my project directory?" Machine: Using execute-command to run npm install... ``` ### create-file Create a new file with specified content. ``` You: "Create a new HTML file for our landing page" Machine: Using create-file to generate index.html... ``` ### delete-file Remove a file from the file system. ``` You: "Delete the temporary log file we created" Machine: Using delete-file to remove temp.log... ``` ### full-file-rewrite Completely replace the contents of an existing file. ``` You: "Rewrite the CSS file with the updated styles" Machine: Using full-file-rewrite to update styles.css... ``` ### str-replace Replace specific strings or patterns within a file while preserving the rest. ``` You: "Replace all instances of the old company name with the new one" Machine: Using str-replace to update references in the files... ``` ## Browser Automation Tools ### browser-navigate-to Navigate to a specific URL in the browser. ``` You: "Go to github.com" Machine: Using browser-navigate-to to access GitHub... ``` ### browser-click-element Click on a specific element on a webpage. ``` You: "Click the login button" Machine: Using browser-click-element to click on the button... ``` ### browser-input-text Enter text into a field on a webpage. ``` You: "Type my email address into the signup form" Machine: Using browser-input-text to enter your email... ``` ### browser-scroll-down Scroll down on the current webpage. ``` You: "Scroll down to see more content" Machine: Using browser-scroll-down to view more of the page... ``` ### browser-scroll-up Scroll up on the current webpage. ``` You: "Scroll back to the top" Machine: Using browser-scroll-up to return to the top of the page... ``` ### browser-scroll-to-text Scroll the page to a specific text element. ``` You: "Find the section about pricing on this page" Machine: Using browser-scroll-to-text to locate the pricing section... ``` ### browser-select-dropdown-option Choose an option from a dropdown menu. ``` You: "Select 'English' from the language dropdown" Machine: Using browser-select-dropdown-option to change the language... ``` ### browser-get-dropdown-options Retrieve all available options from a dropdown menu. ``` You: "What language options are available?" Machine: Using browser-get-dropdown-options to list all languages... ``` ### browser-send-keys Send specific keyboard inputs to the browser. ``` You: "Press Enter to submit the form" Machine: Using browser-send-keys to press Enter... ``` ### browser-switch-tab Change to a different browser tab. ``` You: "Go to the other tab we opened earlier" Machine: Using browser-switch-tab to switch tabs... ``` ### browser-close-tab Close the current browser tab. ``` You: "Close this tab now that we're done" Machine: Using browser-close-tab to close the current tab... ``` ### browser-drag-drop Perform drag and drop operations on webpage elements. ``` You: "Drag the image into the upload area" Machine: Using browser-drag-drop to move the element... ``` ### browser-wait Wait for a specific condition or time period before continuing. ``` You: "Wait for the page to fully load" Machine: Using browser-wait to pause until loading completes... ``` ## Deployment Tools ### deploy Deploy applications, websites, or other resources to production environments. ``` You: "Deploy our website to production" Machine: Using deploy to publish your site... ``` ## Information Tools ### ask Ask Machine a question or request clarification about a topic. ``` You: "What's the best approach for optimizing this code?" Machine: Using ask to analyze and provide optimization recommendations... ``` ### complete Request Machine to complete a task or process that was previously started. ``` You: "Please finish generating the documentation we started" Machine: Using complete to finish the documentation generation... ``` ### crawl-webpage Extract data and information from a specified webpage. ``` You: "Extract all the product information from this catalog page" Machine: Using crawl-webpage to gather the product details... ``` ### web-search Perform a search for information on the internet. ``` You: "Find recent articles about machine learning advances" Machine: Using web-search to find relevant information... ``` ## Using Tools in Conversations You don't need to explicitly name these tools in your conversations. Simply describe what you want to accomplish, and Machine will automatically select and use the appropriate tools. However, you can reference tool names directly if you prefer a specific approach: ``` You: "Can you use browser-navigate-to to open our company website?" Machine: Using browser-navigate-to to access your company website... ``` The availability of specific tools may depend on your subscription plan. Enterprise users have access to additional custom tools and extended functionality. # Use Cases Source: https://support.myapps.ai/machine/use-cases Real-world applications of Machine # Machine Use Cases Discover how people are using Machine to solve real-world problems and enhance their productivity. These examples demonstrate Machine's versatility across different domains and tasks. ## Research & Analysis **Challenge**: An entrepreneur needed comprehensive market research for a new product launch but had limited resources. **Solution**: Machine gathered data from multiple sources, analyzed market trends, identified key competitors, and generated a detailed report—all through a guided conversation. **Outcome**: The entrepreneur saved weeks of research time and gained insights they might have otherwise missed. **Challenge**: A graduate student needed to review hundreds of academic papers for their dissertation. **Solution**: Machine crawled academic databases, extracted relevant information from papers, identified key themes, and compiled a structured literature review. **Outcome**: The student completed their literature review in days instead of months, with more comprehensive coverage than manual research. **Challenge**: A business analyst needed to find patterns in customer feedback data spanning thousands of entries. **Solution**: Machine processed the data, identified recurring themes, performed sentiment analysis, and created visualizations to highlight key insights. **Outcome**: The analyst identified previously unknown customer pain points and implemented targeted improvements. ## Content Creation **Challenge**: A small business owner needed a professional website but lacked technical skills and budget for a developer. **Solution**: Through conversation, Machine helped design, code, and deploy a complete website, guiding the owner through all necessary decisions. **Outcome**: The business launched a professional website in days, improving their online presence and customer reach. **Challenge**: A marketing team needed to create content for multiple channels with consistent messaging. **Solution**: Machine helped develop a content strategy, generated draft content for different platforms, and provided suggestions for optimization. **Outcome**: The team increased their content output by 300% while maintaining quality and brand consistency. **Challenge**: A consultant needed to create standardized reports for multiple clients with consistent formatting. **Solution**: Machine developed templates, generated content based on data inputs, and created professional PDF outputs. **Outcome**: Report creation time decreased from 3 hours to 30 minutes per client while maintaining quality. ## Data Collection & Analysis **Challenge**: A team needed to regularly extract data from multiple websites and compile it into structured reports. **Solution**: Machine automated the entire process—navigating websites, extracting data, transforming it into the required format, and generating reports. **Outcome**: What used to take 15+ hours per week was reduced to a 30-minute automated process with greater accuracy. ## Workflow Automation **Challenge**: A manager was spending hours daily sorting through hundreds of emails. **Solution**: Machine helped create automated workflows to categorize emails, draft responses to common inquiries, and highlight priority messages. **Outcome**: Email processing time reduced by 70%, allowing more focus on strategic work. **Challenge**: A small business struggled to maintain consistent social media presence across multiple platforms. **Solution**: Machine helped create a content calendar, draft posts, and suggest optimal posting times based on audience analytics. **Outcome**: Social media engagement increased by 45% while reducing time spent on management. ## Learning & Education **Challenge**: A student was struggling to understand complex concepts in physics. **Solution**: Machine created personalized explanations, interactive examples, and practice problems, adapting to the student's learning pace and style. **Outcome**: The student's understanding and test scores improved significantly. **Challenge**: A professional wanted to learn programming but found traditional courses too rigid. **Solution**: Machine created a custom learning path, provided interactive coding exercises, and offered real-time feedback on projects. **Outcome**: The learner acquired practical programming skills faster than through traditional courses. ## Business Operations **Challenge**: A growing company couldn't scale their customer support team fast enough to meet demand. **Solution**: Machine helped create automated responses to common queries, develop a knowledge base, and route complex issues to human agents. **Outcome**: Response times decreased by 80% while customer satisfaction increased. **Challenge**: A business needed to document complex internal procedures for new employee onboarding. **Solution**: Machine interviewed stakeholders, compiled information, and created comprehensive step-by-step guides with visuals. **Outcome**: Onboarding time decreased by 40% and new employees reported greater clarity about processes. ## Your Success Story We'd love to hear how you're using Machine to solve your challenges. Share your use case with us at [rcohen@mytsi.org](mailto:rcohen@mytsi.org), and it might be featured in our next update! These use cases represent actual applications of Machine, though specifics have been anonymized for privacy. Results may vary based on the complexity of tasks and quality of inputs. # Using Machine Source: https://support.myapps.ai/machine/using-machine Step-by-step guide to using Machine's interface # Using Machine This guide walks you through the practical aspects of using Machine's interface to accomplish real-world tasks, using web development as an example. ## Interface Overview Machine's interface is designed to be intuitive and conversation-driven. You'll work primarily through a chat interface where you can: 1. Ask questions and give instructions in natural language 2. View Machine's responses and suggested actions 3. Access workspace files and terminal commands 4. Preview results of your tasks Machine Interface Machine Interface ## Key Components ### Chat Interface The main conversation area where you interact with Machine. Simply type your requests in natural language, and Machine will respond with text, suggested actions, or results. ### Terminal Access Machine can execute terminal commands on your behalf. This is particularly useful for: * Running development servers * Installing dependencies * Executing scripts * Performing system operations Machine Terminal Interface ### File Workspace Access, create, and modify files directly within Machine. The workspace shows: * File structure * File content * File creation and editing options Machine Workspace Interface ## Common Workflows ### Web Development Example Let's walk through a common web development workflow using Machine: #### 1. Setting Up a Project Start by describing what you want to build: ``` You: "I need to create a real estate agency homepage with sections for about, contact information, and listings." ``` Machine will help you set up the project structure and create initial files: ``` Machine: "I'll help you create a real estate homepage. Let's start by setting up the basic project structure with HTML, CSS, and JavaScript files." ``` #### 2. Running a Development Server Machine can start a local development server for you: ``` You: "Can you start a local server so I can preview the site?" ``` Machine will execute the appropriate command: ``` Machine: "Starting a Python HTTP server on port 8000. You can view your site at http://localhost:8000" ``` Machine starting a web server #### 3. Making Content Updates You can ask Machine to make specific updates to your files: ``` You: "Let me update the hero section text to reflect the new agency name." ``` Machine will help you edit the specific file and section: ``` Machine: "I'll help you update the hero section. Which agency name would you like to use?" ``` After providing the information, Machine will make the changes and show you what was updated: Machine editing content #### 4. Real-time Previews As you make changes, you can preview your website in real-time: ``` You: "Show me a preview of the website" ``` Machine will provide a way to view the current state of your project: ``` Machine: "Here's the current preview of your website. You can view it in your browser." ``` Machine showing website preview ## Task-Based Examples ### Managing To-Do Items Machine can help you manage task lists and mark items as complete: ``` You: "Let me update the todo list to mark the final task as complete." ``` Machine updating a todo list ### Sharing Conversations You can easily share your conversations with others: ``` 1. Click the three dots menu in the conversation header 2. Select "Share Conversation" from the dropdown menu 3. Choose your sharing method (link, email, or copy) ``` This allows you to share solutions, workflows, or interesting interactions with colleagues or friends without giving them access to your account. Machine sharing a conversation ### Updating Contact Information Update specific sections of your website or documents: ``` You: "Let me update the contact email to reflect the new agency name." ``` Machine updating contact information ### Modifying Footer and Copyright Text Make consistent branding changes across your project: ``` You: "Update the footer text to reflect the new agency name." You: "Update the copyright text to reflect the new agency name." ``` Machine updating footer text ## Working with Files ### Viewing Workspace Files Machine gives you access to your project files: ``` You: "Show me the workspace files" ``` Machine workspace files ### Editing Specific Files You can ask to edit specific files by name: ``` You: "Edit the index.html file to update the navigation links" ``` Machine will open the file and help you make precise changes. ## Best Practices For the best experience when using Machine: 1. **Be specific with your requests** - Clearly describe what you want to accomplish 2. **Provide context** - Let Machine know the purpose behind your changes 3. **Review suggestions** - Always review Machine's proposed changes before confirming 4. **Use iterative refinement** - Start with basic functionality and refine incrementally 5. **Ask for explanations** - If you're unsure about a suggestion, ask Machine to explain ## Keyboard Shortcuts Machine supports several keyboard shortcuts to enhance your productivity: | Shortcut | Action | | ----------------------------- | ------------------------ | | `Ctrl+Enter` or `⌘+Enter` | Send message | | `Esc` | Cancel current operation | | `Ctrl+/` or `⌘+/` | Show available commands | | `Ctrl+Shift+C` or `⌘+Shift+C` | Copy code block | | `Ctrl+Shift+F` or `⌘+Shift+F` | Format code | Machine's interface is constantly evolving with new features and improvements. If you discover a new way to use Machine effectively, please share your feedback with us! # Models Source: https://support.myapps.ai/models How the Pixio model catalog works and how to choose Pixio carries **over 500 models**. This page explains how the catalog is organized so you can find the right one quickly, instead of scrolling. The complete, always-current list — every model, its inputs, its ranges, and its price — lives in the [API reference](https://beta.pixio.myapps.ai/api-docs) and in the model picker inside [Generate](/generate). This page is the map, not the inventory. ## Organized by family Models are grouped by the **family** that makes them. Pick the family first, then the specific model inside it. Families you'll see: | | | | --------------------- | --------------------------------------------------------------------------------------------------------------- | | **Video** | Kling · Seedance · PixVerse · LTX · Veo · Wan · Luma · Hailuo · Vidu · Pika · Runway · Grok · HeyGen | | **Image** | Flux · Imagen · Recraft · Stable Diffusion · Qwen · Ideogram · Reve · Bria · NVIDIA · Magnific · Krea · HiDream | | **Audio & voice** | ElevenLabs · SongCraft · Hailuo · Mureka | | **3D** | Hunyuan · Tripo · Meshy | | **Lipsync & avatars** | Sync · HeyGen · Veed | ## Organized by capability Every model declares what it converts. This is usually the fastest way to filter: | Capability | What it means | | --------------------------------- | ----------------------------------------------- | | **text-to-image** | Prompt in, picture out | | **image-to-image** | Edit, restyle, or transform an existing picture | | **inpainting** | Replace part of an image, keeping the rest | | **text-to-video** | Prompt in, clip out | | **image-to-video** | Animate a still | | **video-to-video** | Restyle, upscale, or transform a clip | | **text-to-audio** | Music, speech, or sound effects from a prompt | | **audio-to-audio** | Voice conversion and audio transformation | | **lipsync** | Match mouth movement to an audio track | | **text-to-3d** / **image-to-3d** | Generate a 3D mesh | | **3d-rigging** / **3d-animation** | Rig and animate a mesh | | **text-to-svg** | Vector output instead of pixels | Image and video dominate the catalog; text-to-image and image-to-video are the two largest groups by a wide margin. ## What each model costs Pricing is per model, and shown in the picker before you run anything: * **Fixed** — a flat credit cost per run, typical for images * **Per second** — video and audio models often scale with duration * **By resolution or quality** — higher settings cost more on many models See [Credits & Plans](/credits) for how billing works. ## What your plan can reach Access is set per model: * Most of the catalog is available on any paid plan * **Free accounts** reach a large subset — enough to evaluate every category * Some models are **free on specific plans**, running without touching your balance * **Maker Mode** covers a wide set of models under unlimited generations Models your plan can't run are marked in the picker rather than hidden, so you can see what an upgrade would unlock. ## Choosing a model Decide what you're converting — text to image, image to video, video to video. That cuts 500 models to a few dozen. Most families ship a fast/cheap model and a flagship. Draft with the cheap one, finish with the flagship. Two models with the same capability can accept very different controls. Every input has a tooltip — read them before assuming a setting exists. Short duration, lower resolution, one output. Confirm the model understands your prompt before spending real credits. For video, generate the first frame as an image, then chain **Create Video** from it. You get far more control over composition, and you iterate at image prices. See [chained actions](/generate#chained-actions). ## New models The catalog changes often. New models appear in the picker as soon as they're enabled — nothing to install or update. Check the [changelog](https://beta.pixio.myapps.ai/changelog) for what landed recently, and [Model Status](https://beta.pixio.myapps.ai/status/models) if a specific model is misbehaving. ## Custom models Trained your own LoRA? It appears in the form for models that support custom weights. Trigger it with its trigger word in the prompt. Training requires Premium or above — see Model Training in the app. # FAQ & Contact Source: https://support.myapps.ai/pixio-canvas/faq Answers about keys, credits, offline drawing, Apple Pencil, storage, stencils, and support. No. Pixio Canvas has no native account registration. Local drawing works immediately. Optional cloud features use a Pixio API key that you provide from your own Pixio account. No. A production or TestFlight build does not bundle a Pixio key. Every user supplies their own. A key already visible on your iPad came from that device's Keychain. No. Quiet mode, drawing, layers, undo and redo, bundled guides, local canvases, and Photo-to-Stencil work without AI. Whisper, Duet, Generate Stencil, Animate, and Pixio Library are cloud features. No. Finger drawing is supported. Apple Pencil provides a more natural drawing and tracing experience when available. Yes. Local drawing and stencil tools work offline after the app is installed. Pixio AI and Pixio Library require internet access. The app does not sell subscriptions, credits, or API access. AI requests use credits from the connected Pixio account. Cost varies by the live model and request. Photo-to-Stencil converts an image you choose on the iPad and saves it under Custom Stencils without an AI request. Generate Stencil sends your description through Pixio to a compatible image model after permission and can use credits. No. Companion results arrive as separate layers. Your ink stays editable, and new ink drawn while Duet is working remains above the result. The local Library stores Pixio Canvas documents on the iPad. Pixio Library is separate and displays cloud assets associated with the connected Pixio key. Open **Settings → Companion Key → Remove key**. You can also withdraw permission under **Settings → AI Processing**. Neither action automatically deletes local drawings or existing cloud assets. Delete a canvas from the local Library, remove an entry from Custom Stencils, or delete an individual canvas layer. The red trash control clears the current canvas only after confirmation. ## Contact Email **Pixio Canvas Support** at [nkukaj@mytsi.org](mailto:nkukaj@mytsi.org). We aim to respond within 24–72 hours. Include the device, OS and app versions, exact message, and reproduction steps. Attach a screenshot only when it does not reveal sensitive artwork or account information. Never send passwords or a complete Pixio API key. Support will not ask you to publish or email the key. * [Pixio Canvas Support hub](/pixio-canvas/support) * [Troubleshooting](/pixio-canvas/troubleshooting) * [Privacy & Data Handling](/pixio-canvas/privacy) * [Pixio Privacy Policy](https://beta.pixio.myapps.ai/privacy-policy) # Getting Started Source: https://support.myapps.ai/pixio-canvas/getting-started Start drawing locally or connect your own Pixio API key for AI companion features. Pixio Canvas supports iPadOS 18 or later and works with either touch or Apple Pencil. It does not require a Pixio Canvas account. ## First launch You can enter your own Pixio API key on **Connect to Pixio**, or continue with the local drawing tools and add a key later in Settings. **Refine**, **Imagine**, and **Animate** can remain on **Auto**. Pixio Canvas resolves a compatible live model for each role. Advanced users can choose another compatible model from the live catalog. Tap **Enter the studio** after saving a key, or **Skip for now** to continue locally, then complete or close the short walkthrough. Before the first AI request, review the **AI Processing** disclosure. Choose **Allow AI Processing** to enable Pixio AI features or **Not Now** to keep using local tools only. Pick a pen, pencil, or marker from the left dock, choose a color and brush size, and draw with your finger or Apple Pencil. ## Add your Pixio API key Your API key belongs to your Pixio account. You can obtain and manage Pixio access at [beta.pixio.myapps.ai](https://beta.pixio.myapps.ai). 1. Open **Settings** in Pixio Canvas. 2. Find **Companion Key**. 3. Paste the complete key into **Pixio API key**. 4. Tap **Save key**. The key is stored in the iPad Keychain and is not displayed again in the app. It authenticates AI and Pixio Library requests and can use credits from the connected Pixio account. Each person should use their own key. Do not share a key in screenshots, support messages, classrooms, public TestFlight instructions, or source code. ## Use Pixio Canvas without a key Without a key, you can still: * Draw, erase, crop, transform, zoom, pan, and rotate * Create, reorder, rename, lock, hide, and adjust layers * Undo, redo, clear, autosave, and reopen local canvases * Browse and trace all bundled guides * Turn a photo into a reusable stencil on the iPad Quiet mode is the clearest local-only workspace. Whisper, Duet, Generate Stencil, Animate, and Pixio Library need a connected key and network access. ## Remove or replace a key Open **Settings → Companion Key** and choose **Remove key**. Removing it disconnects cloud features but does not delete local drawings, custom stencils, or server-side assets already associated with the Pixio account. If a TestFlight build on your iPad shows a key or local artwork you used before, it is reading that same iPad's Keychain and local Documents storage. Those items are not bundled into the app and are not visible to other testers. Next, choose how the companion behaves in [Quiet, Whisper, and Duet](/pixio-canvas/modes). # Layers, Undo & Redo Source: https://support.myapps.ai/pixio-canvas/layers-history Keep ink, stencils, imported images, and companion turns separate and editable. Pixio Canvas lists the topmost layer first. Your ink is never flattened into a companion result: AI outputs, imported images, and tracing guides remain separate from drawing layers. Editable drawing and tracing-guide layers in Pixio Canvas ## Work with layers Open **Layers** in the upper-right corner. From the panel you can: * Add a drawing layer or tracing stencil * Select the layer that receives new ink * Reorder layers up or down * Adjust visibility and opacity * Lock a reference or guide to prevent accidental changes * Rename or duplicate supported layers * Delete only the selected layer When a companion request starts, Pixio Canvas reserves a result position at that moment. If you continue drawing while it processes, your newer ink stays above the returned result. If your marks appear behind an image or guide, select or add a drawing layer and move it above that layer. ## Undo and redo The arrows at the bottom of the left dock move backward and forward through your editing history in order. * **Undo** removes the most recent change first. * **Redo** restores the next undone change. * If you undo and then make a new edit, that new edit becomes the current timeline and the abandoned redo branch is cleared. * A canceled or deleted in-progress companion turn cannot reappear later as a finished layer. The **…** menu also includes **Undo AI Turn** when a companion pass can be removed without walking through unrelated local edits. ## Clear the canvas The red trash button is grouped with Undo and Redo at the bottom of the left dock. It opens a confirmation before clearing the canvas. **Clear Canvas** is also available from the **…** menu. Pixio Canvas portrait workspace with Undo, Redo, and the red Clear button in the left dock Clear Canvas removes the current canvas contents after confirmation. To remove only one item, delete its layer instead. ## Local Library and Pixio Library * **Library** contains canvases autosaved in Pixio Canvas on this iPad. * **Pixio Library** contains cloud assets belonging to the connected Pixio API-key account. Changing keys changes which account-scoped cloud assets appear, but it does not replace the local canvases stored on the iPad. # Quiet, Whisper & Duet Source: https://support.myapps.ai/pixio-canvas/modes Control when the Pixio companion joins your drawing and what starts an AI request. The mode selector determines when AI is allowed to respond. You can change modes at any time without changing your local artwork. Pixio Canvas creative controls with style, strength, intent, and mode choices | Mode | How it behaves | Starts an AI request? | | ----------- | ----------------------------------------------------------------- | -------------------------------------- | | **Quiet** | Keeps the companion asleep while local drawing remains available. | No | | **Whisper** | Waits until you tap the violet **Spark** button. | Only when you tap Spark | | **Duet** | Responds automatically after your pencil rests. | Yes, after the selected rhythm's pause | Whisper and Duet require a Pixio API key, network access, available credits, and permission under **Settings → AI Processing**. If **Settings → Manual volleys only** is enabled, Duet will not start automatic turns. Use the Spark button or turn that setting off to restore pause-triggered responses. ## Quiet Use Quiet for uninterrupted drawing, offline practice, classroom exercises, or tracing without AI. Spark, Animate, and Generate Stencil appear inactive while Quiet is selected, but local tools, layers, bundled guides, and Photo-to-Stencil remain available. ## Whisper Whisper gives you a manual checkpoint: Finish the portion you want the companion to see. **Faithful** preserves more of your marks. Moving toward **Dream** gives the model more freedom. Type a direction such as “add a glowing sunset” or use the microphone when on-device dictation is available. Pixio Canvas snapshots the relevant artwork and starts one companion turn. ## Duet Duet is the continuous co-drawing mode. Draw, pause, and the companion takes a turn. Choose **Flowing** for a quicker rhythm or **Patient** when you want more time between marks and a response. You can continue drawing while a Duet request is processing. The companion result arrives as a separate layer below any newer ink, so it does not cover or replace the strokes you added while waiting. If you want to control every request, use Whisper. If you want no AI requests at all, use Quiet. ## Styles, strength, and intent * **Style** sets the visual direction, including Storybook, Watercolor, Comic, Pencil, and other presets. * **Strength** balances preservation of your marks against creative freedom. * **Intent** tells the companion what to add, change, preserve, or remove. * **Remove** lets you mark an area you want removed when the selected model supports image editing. * **Animate** turns suitable canvas artwork into a video with the selected Animate model. Available models and their options come from Pixio's live catalog, so model names, completion time, formats, and credit cost can change. ## Voice intent Pixio Canvas uses on-device speech recognition and does not fall back to server recognition. If the current language is not available on-device, type the intent instead. When you use the transcript in an AI action, the resulting text can be included in that request after AI-processing permission is granted. # Privacy & Data Handling Source: https://support.myapps.ai/pixio-canvas/privacy Understand what stays on your iPad, what an AI request sends, and how to change permission. Pixio Canvas asks for explicit permission before sending creative content to Pixio or the independent AI provider serving the model selected in Studio. Local drawing remains available when you choose **Not Now**. The Pixio Canvas AI Processing permission screen ## What each action uses | Action | What can leave the iPad | Destination | | ---------------------------------------------- | ---------------------------------------------------------------------------------- | ------------------------------------------------------------- | | Draw, erase, layers, undo, redo, local Library | Nothing for that action | Stays on the iPad | | Bundled tracing guides | Nothing for that action | Stays on the iPad | | Photo-to-Stencil | Nothing for conversion | Processed and saved on the iPad | | Whisper or Duet refinement | Prompt or speech transcript and a flattened canvas or imported image when required | Pixio API and the provider serving the selected Refine model | | Generate Stencil | Your subject description and stencil instructions | Pixio API and the provider serving the compatible image model | | Animate | Canvas image and optional intent required by the selected model | Pixio API and the provider serving the Animate model | | Pixio Library | Key-authenticated requests for that account's cloud assets | Pixio API | The model provider varies with your live model selection. Provider handling and retention can vary, so review the [Pixio Privacy Policy](https://beta.pixio.myapps.ai/privacy-policy) before enabling AI processing. ## Your API key The key is stored in the device Keychain, is not bundled with Pixio Canvas, and is not displayed again after saving. It is transmitted to Pixio when needed to authenticate a cloud request. AI requests can use credits from that key's Pixio account. Never put a key in artwork, prompts, screenshots, email, shared notes, or public testing instructions. ## AI-processing permission Open **Settings → AI Processing** to review the disclosure, allow processing, or withdraw permission. * **Not Now** prevents new AI requests while local tools continue working. * **Allow AI Processing** enables requests from Whisper, Duet, Generate Stencil, and Animate. In Duet, a request starts automatically after you pause drawing. * Withdrawing permission cancels queued and locally tracked work. A request already accepted by a provider may still finish and may still use credits. ## Photos and voice * Photo access is requested only when you explicitly import an image or save exported artwork. * Pixio Canvas does not use the camera. * Microphone and Speech permissions are requested only when you use optional voice intent. * Dictation requires on-device recognition. Raw microphone audio is not sent to Pixio; the resulting text can become part of an AI request after permission. ## Remove data from the app * Remove the stored API key in **Settings → Companion Key**. * Delete individual canvases from the local Library. * Delete individual entries from **Custom Stencils**. * Use the red Clear button to clear the current canvas after confirmation. Removing a key does not automatically delete local drawings, custom stencils, or existing server-side Pixio assets. For account or cloud-data requests, contact [Pixio Canvas support](mailto:nkukaj@mytsi.org) without including the key itself. Pixio Canvas has no in-app public feed, messaging, social-sharing network, advertising, native account registration, in-app purchases, or purchase links. You can explicitly share an animation through Apple's system Share sheet. # Tracing & Stencils Source: https://support.myapps.ai/pixio-canvas/stencils Practice with built-in guides, convert photos on-device, or generate a custom traceable stencil. Open **Stencil Studio** from the pencil-and-outline control or choose **Layers → Add → Tracing Stencil**. ## Built-in guides Pixio Canvas includes 28 skill-building guides across animals, nature, people, objects, vehicles, and imagination. Search by subject or skill, filter by category, and use the difficulty indicators to move from simple contours to more advanced forms. The Pixio Canvas catalog of built-in tracing guides Open **Guides**, search or filter the catalog, then tap a stencil. Move, resize, and rotate the guide until it fits your composition. Make the linework faint enough to trace comfortably. Keep the guide locked beneath your drawing layer so your marks remain independently editable. When the tracing is complete, turn off the guide's visibility or remove only that layer. ## Photo-to-Stencil on this iPad Use **Custom → Upload Photo** to convert a selected image into clean black-and-white tracing lines. This conversion runs on-device and does not use Pixio AI or credits. A custom photo stencil saved and placed in Pixio Canvas For the clearest result, choose a well-lit image with a distinct subject, simple background, and visible edges. Select **Basic** for essential outlines or **Advanced** for more detail. Name the finished result and save it in **Custom Stencils** for reuse. ## Generate Stencil with Pixio Open **Generate**, describe what you want to trace, choose **Basic** or **Advanced**, then tap **Generate Stencil**. Generate Stencil in Pixio Canvas with a fish prompt Pixio Canvas adds strict stencil instructions to your description so the model produces strong black contours on a white background. Auto prefers Nano-Banana 2 when it is available in the live model catalog and falls back to another compatible image model when needed. Generated stencils require: * Whisper or Duet rather than Quiet * AI-processing permission * Your Pixio API key * Internet access and available Pixio credits Preview the result before placing it. Name the result, then choose **Save Only** to keep it in Custom Stencils or **Save & Place** to save and add it to the canvas. Photo-to-Stencil is local. Generate Stencil is an AI request: your description is sent through Pixio to the provider serving the selected model after you grant permission. # Pixio Canvas Support Source: https://support.myapps.ai/pixio-canvas/support Help for drawing, tracing, layers, Pixio AI features, privacy, and your iPad setup. Pixio Canvas app icon Pixio Canvas is an iPad drawing studio built for finger or Apple Pencil. Draw locally, learn with traceable guides, turn your own photos into stencils on-device, or invite a Pixio-powered companion to refine and animate your work. Drawing, layers, undo and redo, bundled tracing guides, and on-device Photo-to-Stencil work without a Pixio API key. AI features are optional. Start locally or connect your own Pixio API key for companion features. Choose whether the companion stays off, waits for the Spark, or responds after you pause. Use 28 built-in guides, convert a photo locally, or generate a custom guide. Keep every drawing and companion pass editable and move backward or forward safely. See exactly what stays on the iPad and what is sent when you allow an AI request. Resolve key, credit, generation, Duet, photo, stencil, and voice issues. Pixio Canvas drawing workspace on iPad ## Local tools and AI tools | Works locally | Uses Pixio cloud services | | ------------------------------------ | -------------------------- | | Drawing and erasing | Refine with the Spark | | Layers, undo, redo, and Clear Canvas | Automatic Duet turns | | 28 bundled tracing guides | Generate Stencil | | Photo-to-Stencil and Custom Stencils | Animate | | Local canvas Library | Pixio Library cloud assets | Pixio Library needs a connected key and internet access. Refine, Duet, Generate Stencil, and Animate additionally need AI-processing permission and available credits. ## Contact Pixio Canvas support Email [nkukaj@mytsi.org](mailto:nkukaj@mytsi.org) and include: * Your iPad model and iPadOS version * The Pixio Canvas version or TestFlight build number * The mode you were using: Quiet, Whisper, or Duet * What you expected, what happened, and the exact message shown * A screenshot when it does not contain private artwork or account information Never email your Pixio API key. Support does not need the full key to troubleshoot the app. Read the [Pixio Privacy Policy](https://beta.pixio.myapps.ai/privacy-policy) or browse the [frequently asked questions](/pixio-canvas/faq). # Troubleshooting Source: https://support.myapps.ai/pixio-canvas/troubleshooting Fix common API key, credit, Duet, stencil, photo, layer, and voice problems. Start by updating to the newest Pixio Canvas build, confirming that the iPad is online, and reading the message shown by the companion. User-facing errors appear in the studio; more technical diagnostics are available in Settings. Copy the complete key from your own Pixio account, remove leading or trailing spaces, and save it again under **Settings → Companion Key**. Confirm the key is active and belongs to an account with available credits. Never send the key to support. Open **Settings → AI Processing**, review what will be sent and where, then choose **Allow AI Processing** if you want to use the AI feature. Choosing **Not Now** intentionally keeps cloud features disabled while local tools remain available. Quiet turns AI tools off. Switch to **Whisper** or **Duet**, then confirm a key, permission, internet connection, and available credits are present. Confirm **Duet** is selected, finish the current stroke, and pause for the selected **Flowing** or **Patient** rhythm. Also check AI-processing permission, the connected key, network access, credits, and the companion message. Use Whisper and tap Spark if you want to test a manual turn. If **Settings → Manual volleys only** is enabled, automatic Duet turns stay off until you disable it. Pixio Canvas sends one generation at a time. Model queues and processing time vary. Keep the app open while the request is active. If it fails, review the companion message, check your credits and connection, then retry once rather than repeatedly starting new work. Open **Layers** and select the top drawing layer, or tap **Add → Drawing Layer** and move it above the result. Companion images and your ink are separate layers. Ink added while Duet is processing should remain above the returned result. Update to the latest build. Undo and Redo should follow edit order, and drawing after Undo should clear the abandoned redo branch. If the issue continues, note the exact sequence of actions and send it to support without private artwork or keys. Tap the red trash button below Undo and Redo, or open **… → Clear Canvas**, then confirm. To preserve the rest of the canvas, delete only the unwanted layer instead. Choose a well-lit image with one distinct subject and a simple background. Try **Basic** for essential outlines and **Advanced** for more details. Photo-to-Stencil runs locally and does not need credits. Generate Stencil is different from Photo-to-Stencil. It requires Whisper or Duet, AI permission, a connected key, internet, and credits. Use a concrete subject such as “friendly fish, side view” and retry after checking the companion message. Allow Microphone and Speech Recognition in iPad Settings. Voice intent also requires on-device recognition for the current language. If that is unavailable, type the intent; Pixio Canvas does not fall back to server-side dictation. TestFlight uses the same iPad Keychain and local app storage across updates. A previously saved key or local canvas can therefore remain on that device. It is not packaged into the build and other testers do not receive it. Remove the key in Settings and delete local canvases from Library when you want a clean workspace. Pixio Library is scoped to the connected API-key account. It may show earlier uploads and generations made with that same account. Connect a different personal key or manage the assets through the Pixio service; local canvases are separate. ## Still need help? Email [nkukaj@mytsi.org](mailto:nkukaj@mytsi.org) with your iPad model, iPadOS version, Pixio Canvas version, mode, exact error message, and reproduction steps. Do not include the API key. # Adding a Persona Source: https://support.myapps.ai/premium/adding-a-persona-multibot Adding a persona in Multibot ### Introduction to Adding a Persona in MultiBot MultiBot offers a versatile platform for creating and managing multiple chatbots, each with its unique personality and capabilities. Personalizing your chatbot with a specific persona is crucial for enhancing user interaction and ensuring your bot aligns with your brand or intended purpose. Whether you're aiming for a friendly, professional, or quirky chatbot, adding a persona is a straightforward process within MultiBot. For users with AI Tutor Plus, the persona creator feature further simplifies this process, allowing for even more detailed and nuanced personas based on custom options. ### How to Add a Persona in MultiBot * Access Settings: Start by locating the settings icon at the top of the MultiBot interface. This icon is your gateway to customizing your chatbot's settings and persona. ![](https://docs.myapps.ai/storage/page_media/aba0cb82-026f-4275-bce9-89c7d70f8a30.png) * Enter Chatbot Name: Once you've opened the settings, enter the name of your chatbot. This name will be the identity of your bot across interactions, so choose something that reflects its persona and purpose. * Add Your Persona under Custom Instructions: Below the chatbot's name, you'll find the option to add your persona under "Custom Instructions." This is where you can define the characteristics, tone, and behavior of your chatbot. Be as detailed as possible to give your chatbot a distinct and engaging personality. * You can also adjust the "temperature", which determines how on topic your chat bot will be or how off topic it can be. Feel free to play around with the settings until your chat bot responds in the manner you like. ![](https://docs.myapps.ai/storage/page_media/735b0001-0c7a-46a1-91db-e7614518e17a.png) * Using AI Tutor Plus for Persona Creation: If you have access to AI Tutor Plus, take advantage of the persona creator feature. This tool allows you to generate a persona using the "Custom" option, providing a guided approach to defining your chatbot's personality traits, preferences, and interaction style. * Crafting Your Own Prompt: For those without AI Tutor Plus, fear not. You can still come up with your own prompt to add to your chat persona. Think about how you want your chatbot to interact with users, its knowledge base, and any unique quirks or phrases it might use. Incorporate these elements into your custom instructions to bring your chatbot's persona to life. ### Tips for Creating an Effective Chatbot Persona * Define Clear Characteristics: Determine the key traits of your chatbot's persona, such as friendly, authoritative, humorous, or empathetic. These characteristics should guide all interactions. * Consider Your Audience: Tailor your chatbot's persona to match the expectations and preferences of your target audience. A well-aligned persona can significantly enhance user engagement. * Be Consistent: Ensure that your chatbot's persona is consistent across all interactions. Consistency helps build trust and familiarity with users. * Iterate and Improve: Don't hesitate to refine your chatbot's persona based on user feedback and interaction patterns. A persona can evolve over time to better meet user needs. ### Conclusion Adding a persona to your MultiBot chatbot is a crucial step in creating a memorable and engaging user experience. Whether you're leveraging the AI Tutor Plus persona creator or crafting your own custom instructions, defining a clear and consistent persona can transform how users perceive and interact with your chatbot. By following the steps outlined above and considering the tips for creating an effective persona, you can ensure your chatbot stands out and effectively communicates your intended message or brand identity. # Agent Builder Source: https://support.myapps.ai/premium/agent-builder Agent Builder # General Overview #### Introduction Utilizing AI agents effectively requires access to powerful tools and resources. OpenAI's API key is essential for anyone looking to harness the capabilities of AI in various applications, including chatbots and web crawlers. This guide provides a comprehensive overview of how to navigate and use AI Agent through the AI Tutor Premium platform, enhancing your ability to integrate AI functionalities into your projects. #### Getting Started with AI Agent 1. Obtain an OpenAI API Key: * Before you can start using AI Agent, you need to secure an OpenAI API key. This key is crucial as it connects your applications to OpenAI's powerful AI models. You can obtain this key by registering on the OpenAI website and following the necessary steps to generate an API key. 2. Access AI Tutor Premium: * To find AI Agent, visit the AI Tutor Premium page. From there, browse through the list of featured apps and select AI Agent. This will direct you to the homepage of the application. #### Using AI Agent 1. Navigate the Sidebar Menu: * On the homepage, you'll find a left sidebar menu. This menu includes various options such as files and web crawlers, which are instrumental in gathering and organizing information for your AI chat. 2. Add Information to Your Chat: * Web Crawlers: Click on the web crawlers option to add URLs. This feature allows the AI to crawl specified websites and gather relevant information. * Files: If you have documents such as PDFs that contain necessary information, you can upload these by selecting the files option. This uploads and integrates the content into your AI chat. 3. Create and Configure Chatbots: * Navigate to the chatbots section and create a new chatbot. During the setup, you can choose to extract information from either the web crawlers or the files you've added. This step will require your OpenAI API key, so ensure it is readily available. 4. Embedding the Chatbot on Your Website: * Once your chatbot is configured and ready, you might want to integrate it into your website. To do this, click on the three dots next to your chatbot on the right-hand side and select the 'Embed on Website' option. This provides you with the necessary code or widget to add the chatbot seamlessly to your site. #### Conclusion AI Agent, powered by OpenAI's API, offers robust tools for creating intelligent chatbots and utilizing web crawlers to enhance your digital projects. By following this guide, you can effectively navigate the AI Tutor Premium platform, set up your AI functionalities, and integrate them into your websites or other projects. Remember, the key to leveraging AI effectively lies in understanding how to configure and apply these tools to meet your specific needs and objectives. # AI Console Source: https://support.myapps.ai/premium/ai-console AI-powered full-stack web development directly in your browser. # General Overview #### Introduction **AI Console** is an advanced, AI-powered development environment that enables **full-stack web development directly in your browser**. It supports **multiple LLMs** and offers an **extensible architecture** for seamless integration of additional models. With powerful features like **image attachments in prompts, integrated terminals, version control, and Docker support**, AI Console provides a **streamlined coding experience** for developers, researchers, and AI enthusiasts. This guide walks you through the **core features, integrations, and upcoming enhancements** in AI Console. *** ### Key Features #### **1. AI-Powered Full-Stack Development** * Develop and **execute code directly in your browser** using AI-assisted tools. * Supports **multiple programming languages** and frameworks. #### **2. Multi-LLM Support** * AI Console integrates with **various large language models (LLMs)**. * Extensible architecture allows **easy integration of new models**. #### **3. Image Attachments for Contextual Prompts** * Attach **images to prompts** to provide better context to LLMs. * Enhances AI’s **understanding of visual elements** in development. #### **4. Integrated Terminal for LLM Command Execution** * View real-time **command execution output** within the console. * Debug and analyze AI-generated commands **without leaving the interface**. #### **5. Code Version Control & Revert** * Easily **revert code to earlier versions** for debugging. * Enables **quick iteration and error recovery**. #### **6. Project Portability & ZIP Downloads** * Download entire projects as **ZIP files** for easy sharing and portability. * Ensures **seamless migration** between different development environments. #### **7. Docker Integration for Hassle-Free Setup** * AI Console is **containerized with Docker** for easy installation. * Enables **quick deployment** on local machines or cloud environments. *** ### Project Management & Community Contributions **AI Console is a community-driven project!** While the core team organizes development efforts, contributions from the community help shape its future. If you’re interested in contributing, check out the **Project Management Guide** to get started. *** ### Recent & Upcoming Integrations The AI Console team is actively working on **new integrations and enhancements** to improve functionality. Below is a list of requested and implemented features: #### ✅ **LLM & API Integrations** * **OpenRouter Integration** * **Gemini Integration** * **Mistral API Integration** * **Cohere Integration** * **DeepSeek API Integration** * **xAI Grok Beta Integration** * **HuggingFace Integration** * **AWS Bedrock Integration** * **Together Integration** * **Perplexity Integration** #### ✅ **AI Development Enhancements** * **Autogenerate Ollama models** from downloaded files * **Filter models by provider** * **Dynamic model max token length** * **Better prompt enhancing** * **Prompt caching** for improved efficiency * **Attach images to prompts** for better context * **PromptLibrary** for multiple variations of prompts #### ✅ **Code & Project Management** * **Ability to sync files (one-way sync) to local folder** * **Load local projects into the app** * **Git Clone button** for easy repository access * **Git Import from URL** * **Detect `package.json` and auto-install dependencies** * **Selection tool to target changes visually** #### ✅ **Debugging & Terminal Enhancements** * **Bolt terminal** to view LLM command execution * **Streaming of code output** * **Detect terminal errors and ask AI to fix them** * **Detect preview errors and ask AI to fix them** #### ✅ **Deployment & Cloud Features** * **Containerize the application with Docker** for easy installation * **Publish projects directly to GitHub** * **Ability to enter API keys in the UI** * **Add Starter Template Options** for quick project setup #### ✅ **User Experience & Accessibility** * **Mobile-friendly UI improvements** * **Chat history backup & restore functionality** *** ### Best Practices for Using AI Console To maximize your efficiency with **AI Console**, follow these best practices: ✅ **Use structured prompts** to guide AI-generated code.\ ✅ **Attach images** where necessary for better contextual understanding.\ ✅ **Regularly revert and compare code versions** for debugging.\ ✅ **Utilize the integrated terminal** to view AI-run commands in real time.\ ✅ **Leverage Git integration** for version control and collaboration.\ ✅ **Deploy with Docker** for a stable and consistent development environment. *** ### Future Roadmap 🚀 **Expanded LLM support** for more AI models.\ 🚀 **Enhanced AI-driven debugging tools** for better error detection.\ 🚀 **Live collaboration features** for real-time coding with teams.\ 🚀 **More cloud deployment options** for seamless hosting.\ 🚀 **Automated testing integration** for AI-generated code validation. *** ### Conclusion AI Console is a **powerful, AI-driven development environment** that simplifies **full-stack web development** with **real-time AI assistance**. Whether you're a **developer, researcher, or AI enthusiast**, this tool provides a **seamless coding experience** with **multi-LLM support, integrated debugging, and project portability**. By following this guide, you can **effectively leverage AI Console** for **faster, smarter, and more efficient** web development. *** # AI Research Source: https://support.myapps.ai/premium/ai-research AI Research # General Overview #### Introduction AI Researcher is a tool designed to streamline the research process by leveraging artificial intelligence to gather and synthesize information on a wide range of topics. This guide will walk you through the steps to effectively use AI Researcher, making your research efforts more efficient and comprehensive. #### How to Use AI Researcher 1. Select a Research Topic: * Begin by choosing the topic you want to research. You can be as specific or as vague as you like. Specific topics will yield more targeted results, while vague topics may provide broader insights. 2. Choose the Depth of Research: * AI Researcher offers a dropdown menu where you can select how in-depth you want the research to be. Options typically range from a basic overview to a detailed analysis. Choose the level that best suits your needs. 3. Select Resource Preferences: * Decide what kind of resources you would like the AI to utilize during the research process. This could include academic journals, books, websites, and other credible sources. Selecting the right resources ensures that the information gathered is reliable and relevant. 4. Initiate the Research Process: * Once you have set your parameters, start the research process. AI Researcher will begin gathering and analyzing information based on your specifications. The duration of this process can vary from a few moments to several minutes, depending on the complexity and depth of the research. 5. Review the Research Summary: * After the research is complete, AI Researcher will provide you with a concise summary of the findings. This summary includes key points and insights derived from the sources it examined. 6. Citations and Download Options: * The summary will also include citations at the bottom, detailing the sources used during the research. This is crucial for verifying the information and for further reading. * You have the option to download the summary along with the citations as a PDF document. This feature is particularly useful for personal use, academic purposes, or professional reference. #### Conclusion AI Researcher is an invaluable tool for anyone looking to enhance their research efficiency. By allowing users to specify the topic, depth, and resources, it personalizes the research process to meet individual needs. Whether you are a student, a professional, or just curious about a topic, AI Researcher provides a quick and easy way to access organized and credible information. Embrace this AI-powered tool to save time and effort in your research endeavors. # AI Tutor Artifacts Source: https://support.myapps.ai/premium/ai-tutor-artifacts AI Tutor Artifacts # General Overview #### Introduction The AI Tutor Artifact is a groundbreaking premium tool designed to transform the way we think about and create digital applications. By harnessing the power of AI, this tool simplifies the app development process, allowing users to bring their imaginative ideas to life with ease. Whether you're a seasoned developer or a beginner with a vision, AI Tutor Artifact provides the necessary resources to build functional and innovative apps. #### How to Use AI Tutor Artifact 1. Select the Appropriate AI Model: * Next to the chatbot, there is an option to choose the type of AI model you want to use. Select the model that best suits the complexity and requirements of your project to ensure optimal outcomes. ![](https://docs.myapps.ai/storage/page_media/c96efc01-a9ad-4fb3-af3f-966201513bcb.png) 2. Initiate Your Project: * Start by typing into the chat what you would like to create. For instance, if you want to develop a calculator, simply state, "I want to create a functional calculator." Be specific about your requirements to ensure the AI understands your vision. 3. Receive AI-Generated Suggestions: * After submitting your request, you will receive a response that includes a brief summary of your project. Additionally, you'll be presented with two clickable options: one showing the ideal code necessary to build your app and another providing a preview of what the finished product would look like. ![](https://docs.myapps.ai/storage/page_media/302a4f91-2d16-4f79-967e-7143ceaa9e0b.png) For a video tutorial, go to [this link](https://www.youtube.com/watch?v=-giAPogpjGc). 4. Refine and Develop Your Idea: * Use the initial suggestions as a starting point to further refine and develop your app. You can add more details or request adjustments to tailor the project to your specific needs. Remember to specify the programming language you prefer for your code. 5. To start a new chat or revisit previous conversations, click the writing icon at the top left of the page. This helps you manage multiple projects or ideas simultaneously. ![](https://docs.myapps.ai/storage/page_media/78fab66b-8227-4f68-a80a-b3a61207993c.png) If you click on the icon next to this you will see a side bar open up with all your chat history, here you can look through your old chats and previous work as well. ![](https://docs.myapps.ai/storage/page_media/0c59366c-0fda-4343-be24-8fc16a7c40e1.png) 6. Incorporate Media and Additional Resources: * If your app requires media such as images, videos, or documents, you can easily add these by clicking the paperclip button. This feature allows you to integrate various media types directly into your project. #### Conclusion AI Tutor Artifact is set to redefine the landscape of app development by making it more accessible, efficient, and user-friendly. With its intuitive AI-driven interface, users can easily translate their creative concepts into functional applications. Whether you are looking to develop a simple utility app or a complex interactive platform, AI Tutor Artifact equips you with the tools and guidance needed to succeed. Embrace this innovative technology to unleash your imagination and create apps that resonate with your vision and goals. # App Builder Source: https://support.myapps.ai/premium/app-builder App Builder # App Builder Beta Documentation ## Table of Contents 1. [Overview](#overview) 2. [Getting Started](#getting-started) 3. [Choosing a Persona](#choosing-a-persona) 4. [Selecting a Language Model](#selecting-a-language-model) 5. [Configuring Parameters](#configuring-parameters) 6. [Viewing Code and Preview](#viewing-code-and-preview) 7. [Deploying to the Cloud](#deploying-to-the-cloud) 8. [Troubleshooting](#troubleshooting) 9. [FAQ](#faq) ## 1. Overview App Builder Beta is a powerful tool designed to streamline the process of creating applications. With an intuitive interface and advanced features, it allows developers of all skill levels to build, preview, and deploy applications with ease. ## 2. Getting Started To begin using App Builder Beta, navigate to the Premium dashboard in AI Tutor and select the application to get started. There is no way for you to save your chats and a simple refresh can reset the conversation. ## 3. Choosing a Persona One of the unique features of App Builder Beta is the ability to choose a persona for your project. ![Choosing a Persona](https://img.mytsi.org/i/qlhe733.png) * Click on the default option "Auto" above the chat prompt and that will open the dropdown menu * Select from a variety of pre-defined personas, each tailored to specific project types * Your chosen persona will guide the AI in generating more relevant and context-appropriate code ## 4. Selecting a Language Model App Builder Beta integrates with various Language Models (LLMs) to enhance your development experience. ![Selecting a LLM](https://img.mytsi.org/i/FJGv728.png) * Navigate to the default model "Claude Sonnet 3.5" section next to personas * Choose from the available LLMs based on your project requirements * The selected LLM will be used for code generation and chat interactions ## 5. Configuring Parameters Customize your project by configuring various parameters to suit your needs. ![Configuring Parameters](https://img.mytsi.org/i/0beo727.png) * Access the "Parameters" section * Adjust settings such as project type, target platform, and specific features * These parameters will guide the AI in generating appropriate code and structure for your project ## 6. Viewing Code and Preview App Builder Beta offers two main views for your project: Code and Preview. ### 6.1 Code View ![Code View](https://img.mytsi.org/i/lWKT736.png) * Access the generated code for your project * Options to copy code snippets or download entire files * Syntax highlighting for easy readability ### 6.2 Preview ![Preview](https://img.mytsi.org/i/wgD2735.png) * Quickly view a live preview of your project * Interact with your application in real-time * Test functionality and design without leaving the platform ## 7. Deploying to the Cloud App Builder Beta simplifies the deployment process with its built-in cloud deployment feature. ### 7.1 Initiating Deployment ![Deploy to Cloud Button](https://img.mytsi.org/i/q9gS731.png) * Click the "Deploy to Cloud" button to start the deployment process ### 7.2 Setting Expiration Date ![Set Expiration Date](https://img.mytsi.org/i/qYg5732.png) * Choose an expiration date for your deployed application * This feature helps manage resources and ensures your deployments are up-to-date ### 7.3 Finalizing Deployment * Review the deployment settings * Click "Accept and Deploy" * Once deployment is complete, you'll receive a URL to access your live application ## 8. Troubleshooting If you encounter any issues while using App Builder Beta, please refer to our comprehensive troubleshooting guide or contact our support team. ## 9. FAQ Find answers to commonly asked questions about App Builder Beta, its features, and best practices for application development. *** # BotCraft Pro Adding a Rest API Source: https://support.myapps.ai/premium/bocraft-pro-adding-a-rest-api BotCraft Pro Adding a Rest API # Adding a Rest API Let's use the OpenWeatherMap API as an example. This is a popular weather API that provides current weather data for any location. ### Example: OpenWeatherMap API API Endpoint: `https://api.openweathermap.org/data/2.5/weather` Method: `GET` Headers: ``` json ``` ``` { "Content-Type": "application/json" } ``` Query Parameters: * `q`: The city name (e.g., `London`) * `appid`: Your API key from OpenWeatherMap * `units`: Units of measurement (optional, e.g., `metric` for Celsius) ### Example Request URL: ``` https://api.openweathermap.org/data/2.5/weather?q=London&appid=YOUR_API_KEY&units=metric ``` Headers: ``` json ``` ``` { "Content-Type": "application/json" } ``` Body (JSON): ``` json ``` ``` {} ``` ### How to Set Up in Botcraft Pro 1. Select Data Source: REST API 2. Method: GET 3. URL: `https://api.openweathermap.org/data/2.5/weather?q=London&appid=YOUR_API_KEY&units=metric` 4. Headers: ``` json ``` ``` { "Content-Type": "application/json" } ``` 5. Body (JSON): ``` json ``` ``` {} ``` 6. Chat Model: Choose your preferred model (e.g., GPT-3.5 Turbo) 7. Embedding Model: Choose your preferred embedding model (e.g., text-embedding-ada-002) 8. Click Create ### Example Response A successful response from the OpenWeatherMap API will look something like this: ``` json ``` ``` { "coord": { "lon": -0.1257, "lat": 51.5085 }, "weather": [ { "id": 800, "main": "Clear", "description": "clear sky", "icon": "01d" } ], "base": "stations", "main": { "temp": 15.5, "feels_like": 14.8, "temp_min": 14.0, "temp_max": 17.0, "pressure": 1015, "humidity": 72 }, "visibility": 10000, "wind": { "speed": 3.09, "deg": 240 }, "clouds": { "all": 0 }, "dt": 1624531200, "sys": { "type": 2, "id": 2019646, "country": "GB", "sunrise": 1624508400, "sunset": 1624564800 }, "timezone": 3600, "id": 2643743, "name": "London", "cod": 200 } ``` Replace `YOUR_API_KEY` with your actual API key from OpenWeatherMap. If you don't have an API key, you can sign up for one [here](https://home.openweathermap.org/users/sign_up). This setup will allow your bot to fetch and display current weather information for the specified location. # Bookify Advanced Source: https://support.myapps.ai/premium/bookify-advanced Bookify Advanced # Bookify Advanced: Comprehensive User Guide Introduction Bookify Advanced is a powerful tool within the AI Tutor platform that allows you to write full books using advanced AI technology. This guide will provide an in-depth explanation of how to use the application based on the interface shown in the image. Accessing Bookify Advanced 1\. Log into your AI Tutor account. 2\. Navigate to the Bookify Advanced feature. 3\. Ensure you're using the advanced mode, which offers additional features compared to the streamlined version. Interface Overview The interface is divided into several key sections: 1\. Model Selection 2\. Book Creation Process 3\. Topic Definition 4\. Additional Instructions 5\. Writing Style and Complexity Settings 6\. Seed Content 7\. File Upload Let's explore each of these in detail. 1\. Model Selection At the top of the interface, you'll see "Select AI Models" with a notification: "Advanced Mode is in beta: You're using a version with experimental features." This indicates that you're using cutting-edge AI technology that may include new, experimental capabilities. 2\. Book Creation Process The book creation process is divided into three main parts, each with its own AI model selection: a) For creating book title: * Model: llama3-70b-8192 (default) * You can change this by clicking on the dropdown menu if other options become available. b) For creating book structure: * Model: llama3-70b-8192 (default) * Again, changeable via dropdown if needed. c) For creating book content: * Model: llama3-70b-8192 (default) * Same dropdown option for potential changes. 3\. Topic Definition In the main content area, you'll see a section labeled "Topic" with the instruction: "What do you want the book to be about?" * Enter your desired book topic in the text field provided. * Be as specific and detailed as possible to guide the AI in generating relevant content. 4\. Additional Instructions Below the topic, there's an "Additional Instructions" section: "Provide any specific guidelines or preferences for the book's content" * Use this field to give more detailed instructions about what you want in your book. * In the example shown, the user has provided specific instructions for a strength and conditioning book: "This is a strength and conditioning book for people forty years old and up. I want the book to be in-depth, written so anybody 18-90 can understand, providing tangible information, loaded with useful, relatable information, exercises, exercise routines for each muscle group and body part, from head to toe." 5\. Writing Style and Complexity Settings There are two important settings you can adjust: Writing Style: * A dropdown menu allows you to select the writing style. * In the image, it's set to "Casual", but you can likely choose other styles like formal, academic, etc. Complexity Level: * A slider allows you to set the complexity of the content. * It ranges from "Beginner" to "Expert" with several levels in between. * Adjust this based on your target audience's knowledge level. 6\. Seed Content This section allows you to provide initial content or ideas for the AI to work with: "Provide any existing notes or content to be incorporated into the book" In the example, the user has provided detailed seed content about food: "When it comes to cooking delicious and nutritious meals at home, there are many great options to consider. Start with versatile proteins like chicken, fish, or tofu, which can be prepared in countless ways. Incorporate a variety of colorful vegetables such as bell peppers, broccoli, and carrots for added nutrients and flavor. Whole grains like quinoa or brown rice make excellent bases for many dishes. For quick and easy meals, stir-fries, one-pot pastas, or sheet pan dinners are excellent choices. Don't forget to experiment with herbs and spices to enhance flavors without adding excess calories. Lastly, cooking methods like roasting, grilling, or steaming can help retain nutrients while creating delicious meals. Remember, the key to good home cooking is balancing nutrition, flavor, and convenience to suit your lifestyle and preferences." Use this section to: * Provide specific examples * Outline key points you want covered * Give any other relevant information that will help shape the book's content 7\. File Upload At the bottom of the interface, there's a file upload section: "Upload a text file with your seed content (optional)" * You can drag and drop a file here or click to browse your files. * This is useful if you have extensive notes or content in a separate document that you want to incorporate. Generating Your Book Once you've filled out all the necessary sections: 1\. Review your inputs to ensure everything is as you want it. 2\. Click the "End Generation and Download Book" button at the top of the interface. 3\. The AI will then process your inputs and generate your book. 4\. Once complete, you'll be able to download and review your AI-generated book. Additional Notes * The interface mentions "Visit here to use the streamlined version." This suggests there's a simpler version of Bookify available if you don't need all the advanced features. * Remember that while this tool is powerful, it's always important to review and edit the AI-generated content to ensure it meets your standards and accuracy requirements. By following these detailed steps and utilizing all the features Bookify Advanced offers, you can create comprehensive, tailored books on a wide range of topics using cutting-edge AI technology. # Bookify Basics Source: https://support.myapps.ai/premium/bookify-basics Bookify Basics # Bookify Basics Introduction Bookify is a cutting-edge feature available in AI Tutor Premium that harnesses the power of advanced Large Language Models (LLMs) to generate entire books from a single prompt. Using a sophisticated combination of Llama3-8b and Llama3-70b models, Bookify creates comprehensive, structured content on a wide range of topics within seconds. Features 1. Instant Book Generation: Create full-length books on any topic with just a one-line prompt. 2. Adaptive Content: Easily modify the language complexity or add specific requirements, and watch as the entire book adapts to your needs. 3. Code Integration: For technical topics, Bookify can incorporate relevant code examples throughout the chapters. 4. Intelligent Model Switching: Bookify strategically alternates between Llama3-70b and Llama3-8b models to optimize for both quality and speed. 5. Aesthetic Formatting: Generated books feature markdown styling, including tables and code blocks, for a polished presentation. 6. Download Options: Save your generated book as a text file or a styled PDF for offline reading or further editing. 7. Streamlined Interface: User-friendly design within the AI Tutor Premium platform for effortless book creation. How to Use Bookify 1. Access: Log into your AI Tutor Premium account and navigate to the Bookify feature. 2. Enter Prompt: Type your desired book topic or concept into the provided input field. 3. Generate: Click the "Generate Book" button to initiate the book creation process. 4. Review: Once generation is complete, review the book content displayed on the screen. 5. Refine (Optional): If needed, adjust your prompt or add specific requirements to fine-tune the book's content. 6. Download: Choose to download your book as a text file or a formatted PDF. Example Use Cases 1. Educational Materials: Quickly generate textbooks or study guides on specific subjects. 2. Technical Documentation: Create comprehensive manuals or guides for software, products, or processes. 3. Creative Writing Prompts: Use Bookify to generate story outlines or world-building guides for fiction writing. 4. Research Summaries: Compile overviews of complex topics for academic or professional use. 5. Self-Help and Personal Development: Generate books on various life skills, productivity techniques, or personal growth topics. Limitations While Bookify is a powerful tool, users should be aware that: * Generated content may occasionally contain inaccuracies or placeholder information. * The tool is best suited for non-fiction topics; fiction generation is a feature in development. * Content should be reviewed and fact-checked before use in professional or academic settings. Conclusion Bookify represents a revolutionary approach to content creation, offering AI Tutor Premium users the ability to generate entire books in seconds. Whether for personal learning, professional development, or creative projects, Bookify opens up new possibilities in rapid, customized book creation. # BotCraft Adding Sources in BotCraft Pro Source: https://support.myapps.ai/premium/botcraft-add-source BotCraft Adding Sources in BotCraft Pro ### Adding a New Data Source in BotCraft Pro Integrating new data sources into BotCraft Pro enhances your bot's knowledge base, making it more versatile and capable of handling a wide range of queries. Here's a comprehensive guide on how to add different types of data sources to your bot: ### Step-by-Step Guide 1. Access the Data Source Menu: * Navigate to the sidebar and click on the data source icon, which resembles several stacked ovals. 2. Initiate Adding a New Source: * Click on the blue button located on the right side of the screen to add a new data source. 3. Choose Your Data Source Type: * A small window will appear with a list of source options. Here are the available choices: ### Types of Data Sources 1. Webpage: * Description: Link a webpage for your bot to gather information from. * How to Use: Input the webpage URL in the provided field. 2. Text: * Description: Manually insert text information. * How to Use: Type or paste the text into the input box. 3. File: * Description: Upload documents or files for your bot to process. * How to Use: Add files in compatible formats as listed in the input box. 4. Crawler: * Description: Use a web crawler to index website content automatically. * How to Use: Enter the website URL, set the "max depth of the website to crawl," and the "max links to crawl." 5. GitHub: * Description: Link your GitHub repository. * How to Use: Input the GitHub repo URL, specify the branch, and indicate if it is a private repository. 6. YouTube: * Description: Gather information from a YouTube video. * How to Use: Input the YouTube video URL and select the YouTube mode at the bottom. 7. REST API: * Description: Integrate REST API data. * How to Use: Choose between GET or POST, then enter the REST API URL, along with the Header and Body (JSON). 8. Sitemap: * Description: Use a sitemap to provide structured information about your site. * How to Use: Enter the sitemap URL. ### Processing and Testing 1. Processing State: * Once you add a source, it will appear on your data source page and enter a processing state. This may take a few minutes depending on the size and number of data sources. 2. Completion: * The status will update to "finished" once the data is fully processed. 3. Testing Your Bot: * After processing is complete, test your bot to ensure the new data has been successfully integrated and is functioning as expected. ### Conclusion Adding new data sources in BotCraft Pro is a straightforward process that significantly enhances your bot's capabilities. By following these steps, you can seamlessly integrate various types of data, ensuring your bot is well-equipped to handle a diverse array of queries and tasks. Whether linking a webpage, uploading files, or integrating a REST API, BotCraft Pro provides the tools necessary to create a robust and knowledgeable bot. # BotCraft Pro Bot Setup Source: https://support.myapps.ai/premium/botcraft-bot-setup BotCraft Setting up a bot ### Introduction to BotCraft Pro BotCraft Pro offers a comprehensive platform for creating and deploying sophisticated bots tailored to a variety of needs and functions. From gathering and processing information to interacting with users across multiple channels, BotCraft Pro equips users with the tools necessary to design, build, and refine intelligent bots. ### Creating a New Bot in BotCraft Pro 1. Initiate Bot Creation: Begin by creating a new bot. You'll be presented with a pop-up screen that prompts you to input various forms of information to educate and inform your bot. ![](https://docs.myapps.ai/storage/page_media/b9bbcd7f-8564-422b-9c95-4460df28e041.png) 2. Feed Your Bot: Select and upload the desired files, videos, texts, or website links that will serve as the knowledge base for your bot. This information will be crucial in shaping your bot's responses and actions. 3. Creation and Testing: Once you've inputted the initial information, click create to finalize your bot. After creation, take the opportunity to test your bot, ensuring it processes and responds as expected. ### Utilizing BotCraft Pro Tools ![](https://docs.myapps.ai/storage/page_media/1a78e6cc-a53b-464e-ab10-a4e3e15962ed.png) * Data Sources: This section allows you to view, evaluate, and augment the data feeding your bot's brain. Adding more information over time can help improve your bot's accuracy and functionality. * Integrations: Navigate to integrations to make your bot available on various channels and platforms, expanding its reach and accessibility to users wherever they are. * Conversations: Here, you can monitor all interactions with your bot, gaining insights into user queries and your bot's responses, which can be invaluable for further refinement. * Appearance: Customize the look of your bot by adjusting colors and themes in the appearance section, ensuring your bot aligns with your brand or personal aesthetic preferences. * Settings: For more nuanced control over your bot's behavior and capabilities, explore the settings tools. This area houses advanced settings that can significantly impact your bot's performance. ### Tips for Success with BotCraft Pro * Diversify Your Bot's Knowledge Base: Regularly update and expand the information you provide to your bot to ensure it remains relevant and capable of handling a wide range of queries. * Leverage Integrations: Make the most of BotCraft Pro's integration capabilities to ensure your bot is accessible across the platforms your audience frequents. * Monitor and Refine: Use the Conversations tool to identify areas where your bot may need improvement and adjust your bot's data and responses accordingly. * Personalize Your Bot: Utilize the Appearance settings to give your bot a unique and engaging look that resonates with users. ### Conclusion BotCraft Pro is a powerful tool for anyone looking to create and deploy intelligent bots. By carefully curating the information fed into your bot, leveraging the platform's extensive tools, and continuously refining your bot based on user interactions, you can create a bot that not only performs tasks efficiently but also enhances user engagement. Whether for business, entertainment, or personal assistance, BotCraft Pro provides the resources and flexibility needed to put your bot on the map. # BotCraft Pro Source: https://support.myapps.ai/premium/botcraft-pro BotCraft Pro ## Overview BotCraft Pro is an advanced chatbot creation and management platform that allows users to build, customize, and integrate AI-powered chatbots across various channels. This platform provides a comprehensive suite of tools for users to create chatbots with customizable flows, themes, settings, and integrations. ## User Interface ### Main Navigation * Sidebar Menu: Access different sections of the platform like Playground, Data Sources, Integrations, Conversations, and Appearance. * Top Bar: Contains access to bot settings and user account details. ### Playground * Chat Interface: Simulates the chatbot conversation in real-time. * Settings Gear: Opens the Playground settings where you can manage features like Text to Speech. ### Data Sources * Add New Source: Import content to be used by the bot for providing responses. * List of Sources: View and manage the data sources that the bot currently uses. ### Integrations * Available Channels: Connect the bot to different platforms such as Telegram, Discord, and WhatsApp. * API: Customize the integration using the robust API to extend the chatbot's capabilities. ### Conversations * History: Review past interactions and conversations handled by the bot. ### Appearance * Customization Options: Modify the look of the chatbot, including the bot's greeting message, bubble styles, and color schemes. ### Creating a Bot * Data Source Selection: Choose the type of data source for the bot to use, like webpages, text, or files. * Chat Model: Select the AI model, like GPT-3.5 or GPT-4, to power the chatbot's conversations. * Theme Customization: Pick or customize a theme that matches your brand or preference. ### General Settings * Project Name: Assign a name to your chatbot project. * Model Configuration: Choose the chat model and set parameters such as temperature and streaming. * Prompt Settings: Customize the prompts for the chatbot. * Deletion: Securely delete your bot, along with all its data. ## Functionalities ### Chat Simulation * Simulate real-world interactions with the chatbot to test its responses and flows. ### Custom Flows * Create unique conversation paths based on user input, including conditions, actions, and redirects to guide the user through a seamless conversational experience. ### Theme Editor * Personalize the visual aspects of your chatbot interface, such as background colors, text colors, and bubble styles. * Adjust the bot and user avatars, input fields, buttons, and corner roundness for a cohesive look. * Implement custom CSS for advanced styling needs. ### Settings Configuration * Toggle features like typing emulation, user memory, and source citation within the chat. * Optimize the bot's response times and conversation style with adjustable parameters. * Manage metadata settings, including the bot's icon and preview image. ### Integrations Management * Seamlessly connect your chatbot to multiple platforms with simple toggle switches. * Configure platform-specific settings to ensure compatibility and functionality across different mediums. ### Conversation Tracking * Monitor and track user conversations with the bot to gain insights and improve interaction quality. * Filter conversations by channel or status for efficient management. ### Appearance Customization * Set the initial greeting message to create a welcoming first impression for users. * Customize the style of the chat interface to align with the bot's personality and your brand identity. ### Bot Creation * Select a data source or upload content that the bot will use to learn and provide information. * Choose the chat and embedding models to tailor the bot's response behavior and embedding capabilities. ### Deletion and Privacy * Securely remove a bot and all its associated data with a straightforward deletion process. * Ensure privacy and control over your bot's information and user interactions. # BotCraft Pro Settings Source: https://support.myapps.ai/premium/botcraft-pro-settings BotCraft Pro Settings ### Configuring BotCraft Settings BotCraft offers a range of customizable settings to tailor your bot to your specific needs and preferences. Here's a detailed guide on how to navigate and adjust these settings: ### Key Settings and Actions 1. Change Project Name: * Description: The first action you can perform is renaming your project to better reflect its purpose or content. * How to Use: Simply enter the new name in the provided field. 2. Select Your Model: * Description: Choose the AI model that your bot will use. Ensure this model matches your project requirements. * How to Use: Select the desired model from the dropdown menu. 3. Toggle Streaming: * Description: Control whether your bot is live or offline. * How to Use: Switch streaming on or off as needed. 4. Adjust Temperature: * Description: The temperature setting controls the intensity and creativity of the AI's responses. * How to Use: Move the slider to increase or decrease the temperature based on your preference. 5. Edit Embedding Method: * Description: Choose the embedding method for your bot's data processing. If you change this setting, ensure to re-fetch the data source or select a model with matching dimensions. * How to Use: Select the embedding method from the options provided. 6. Set Number of Documents to Retrieve: * Description: Define how many documents the bot should retrieve to generate responses. * How to Use: Enter the desired number in the input field. 7. Adjust Semantic Search Similarity Score: * Description: This score dictates how many documents are considered when generating responses. * How to Use: Modify the similarity score to refine the bot's search results. 8. Customize Prompt Questions: * Description: Personalize the questions your bot asks for prompting. Ensure to keep the last line intact: `{context} Question: {question} Helpful answer in markdown:`. * How to Use: Edit the prompt question text while retaining the embedded code. 9. Edit Follow-Up Question Response: * Description: Customize the bot's follow-up question responses similarly to the prompt questions. * How to Use: Modify the response text, keeping the necessary code intact. 10. Enable Hybrid Search Retrieval: * Description: Use hybrid search retrieval instead of the default semantic search. Note: This works only on the playground UI. * How to Use: Turn on the hybrid search retrieval option. 11. Activate Public Bot Protection: * Description: Protect your public bot from misuse by activating session-based protection. * How to Use: Enable the public bot protection setting. 12. Add Chat Model API Key: * Description: If you have a chat model API key, you can input it here. If not, you can leave this field blank. * How to Use: Enter the API key in the designated field. ### Conclusion Customizing your BotCraft settings allows you to fine-tune your bot's behavior, ensuring it operates according to your specific needs. From adjusting the AI's response intensity to configuring search methods and embedding techniques, these settings provide the flexibility required to create a highly functional and tailored bot. By following this guide, you can effectively manage and optimize your bot's performance, making it a valuable asset for your projects. # Character AI Source: https://support.myapps.ai/premium/character-ai Character AI # Character AI ## Introduction Character AI is an interactive platform that allows users to create and interact with AI-driven characters. Whether for educational purposes, entertainment, or creative writing, Character AI brings characters to life through conversation. ## Key Features * Character Selection: Users can browse and select from a variety of AI characters. * Interactive Chat: Engage in real-time conversations with AI characters. * Custom Character Creation: Users can create their own AI characters with unique personalities and backstories. ## User Interface ### Home Screen * Search Bar: Look up characters or topics using the search function. * Category Tabs: Easily navigate through different character categories, such as "Famous People" or "Philosophy". * Character Cards: Visual representations of each character with a brief description and interaction button. ### Chat Interface * Chat Window: A conversational interface where users interact with the selected AI character. * Input Field: Type your messages to the AI character here. * Send Button: Click to send your message and receive a response from the AI character. * Conversation History: Scroll through the chat history to review past interactions. ### Create Character Page * Name Field: Assign a name to your custom AI character. * Description Field: Provide a short description or role for the AI character. * Category Dropdown: Select a category that best fits your AI character. * Instructions Field: Input detailed instructions for the AI's behavior and conversation style. * Example Conversation: Provide sample dialogues to guide the AI's interactions. * Create Button: Finalize and bring your custom AI character to life. ## Getting Started with Character AI 1. Select a Character: Choose an existing AI character or create a new one. 2. Initiate Chat: Start conversing with the AI character to explore its personality and knowledge. 3. Customize Your Character: Provide specific instructions and dialogues to tailor the AI responses. ## Best Practices * Explore various characters to understand the breadth of interactions available. * Use detailed backstories and instructions to ensure the AI character behaves as intended. * Regularly interact with your characters to refine their conversational capabilities. ## Support and Resources * Tutorial Guides: Access how-to guides for creating and interacting with AI characters. * Community Forum: Share experiences, tips, and custom characters with other users. * Customer Support: Get assistance with any queries or issues related to the application. ## Conclusion Character AI offers a unique and engaging way to experience artificial intelligence. Through creating and interacting with diverse AI personalities, users can enjoy meaningful and entertaining conversations, making Character AI a versatile tool for a wide range of applications. # Datavibes Pro Source: https://support.myapps.ai/premium/datavibes-pro Datavibes Pro # Exploring DataVibes Pro ### Introduction Datavibes Pro is a multifaceted application designed to provide users with a wide range of capabilities, from financial analysis to shopping and beyond. This comprehensive guide will help you navigate through the various features of Datavibes Pro, ensuring you make the most out of its robust functionalities. #### Key Points 1. Financial Analysis: * Querying Stocks: You can ask detailed questions about stocks by entering queries such as "What are Starbucks stocks?". * Stock Graph: The application will present a graph depicting the stock's performance over time. * TradingView Integration: For more in-depth analysis, click on the TradingView button located at the bottom left corner of the stock screen. This will redirect you to TradingView, where you can perform advanced technical analysis. ![](https://docs.myapps.ai/storage/page_media/21d0e870-0e03-4cca-ab4b-7ec1928002ef.png) * Cited Sources: Directly below the stock graph, you will find cited sources that provide additional context and information. Clicking on these sources will take you to the original content for further reading. * Summary and Follow-Up Questions: The response area will include a summary of the stock information along with suggested follow-up questions to help you continue exploring related topics. * Additional Media: On the right side of the screen, you will see relevant YouTube videos and images that you can click on for further viewing, enhancing your understanding of the stock. 2. Shopping: * Local Shopping Queries: Datavibes Pro can assist you in finding products locally. For instance, if you are looking for shoes, simply type your query into the prompt. * Store Map: The application will generate a map showing nearby stores that sell the item you are looking for. * Product Suggestions: To the right of the map, you will find images, videos, and links to various product suggestions. * Direct Purchase Links: If you find an item that interests you, clicking on it will take you directly to the shopping site where you can purchase the item. 3. User Interface and Experience: * Intuitive Design: Datavibes Pro is designed with user experience in mind. The intuitive interface ensures that you can easily navigate through its various features. * Real-Time Updates: Receive real-time updates on stock prices, market news, and other critical information. #### Conclusion Datavibes Pro is a powerful and versatile platform with an extensive range of features, from financial analysis and investment advice to personal finance management and educational resources. By following the outlined steps and exploring the various functionalities, you can leverage Datavibes Pro to make informed decisions, manage your finances effectively, and enhance your overall financial literacy. Dive into the capabilities of Datavibes Pro and unlock a world of opportunities. # New Image Generators in Multibot Source: https://support.myapps.ai/premium/image-gen-multibot Genertae Images with DALLE. ### Introduction to New Image Generation Features In the ever-evolving landscape of AI-driven design and content creation, new tools are constantly being developed to enhance productivity and creativity. Among these innovations are features focused on image generation, designed to streamline various tasks from data visualization to branding. Accessible via the cube icon on the right side of the interface, these tools, including AI Tutor Developer, Logo GPT, and Vector Builder, offer specialized functionalities to cater to specific needs. Whether you're a developer looking to visualize data, a startup in need of a unique logo, or a designer seeking quick vector images, these tools provide efficient solutions to bring your ideas to life. ### Overview of New Image Generation Features #### AI Tutor Developer * Function: Creates charts and visualizations based on data inputs. * Use Case: Ideal for developers, data analysts, and content creators who need to convert complex datasets into clear, comprehensible visual formats. ![](https://docs.myapps.ai/storage/page_media/373eaef4-e3c4-411d-8e4d-c8508c87b776.png) #### Logo GPT * Function: Generates logos based on textual prompts describing brand identity, style preferences, and specific design elements. * Use Case: Perfect for startups, businesses, and individuals looking for quick, AI-generated logos that align with their branding strategy without the need for extensive design experience. ![](https://docs.myapps.ai/storage/page_media/1d2025c9-770d-4029-9cf9-8e45c4b8c3a6.png) #### Vector Builder * Function: Produces vector images based on prompts, offering scalable graphics suitable for various applications without loss of quality. * Use Case: Useful for graphic designers, marketers, and content creators needing high-quality vector images for digital and print media, branding, or promotional materials. ![](https://docs.myapps.ai/storage/page_media/9dc660c8-5728-4204-b006-f3d8e70e431a.png) ### How to Use These Image Generation Features 1. Access the Tools: Click on the cube icon located on the right side of the interface to open the menu of new image generation features. ![](https://docs.myapps.ai/storage/page_media/b6e38909-f3ba-4b87-885b-ecb116958c07.png) 1. Select Your Tool: Choose the tool that best fits your current need---AI Tutor Developer for data visualization, Logo GPT for logo creation, or Vector Builder for vector images. 2. Input Your Prompt: For Logo GPT and Vector Builder, enter a detailed prompt describing what you envision. For AI Tutor Developer, input or upload the data you wish to visualize. 3. Generate Your Image: Initiate the generation process. The tool will use AI algorithms to interpret your prompt or data and create the desired output. 4. Review and Refine: Evaluate the generated image. These tools often allow for adjustments or refinements based on initial results. 5. Download and Use: Once satisfied with the outcome, download the image for your specific use case, whether it's incorporating it into a website, including it in marketing materials, or any other application. ### Tips for Maximizing the Impact of These Tools * Be Specific with Prompts: The more detailed your description, the better the AI can match your expectations, particularly for Logo GPT and Vector Builder. * Understand Your Data: For AI Tutor Developer, a clear understanding of your data will help you choose the most appropriate type of chart or visualization. * Experiment: Don't hesitate to try different prompts or data sets to explore the range of possibilities these tools offer. * Incorporate Feedback: Use feedback from colleagues or your target audience to refine the generated images, ensuring they effectively meet your objectives. ### Conclusion The new image generation features on the platform are powerful tools designed to assist in various creative and analytical endeavors. By leveraging AI Tutor # Magic AI Source: https://support.myapps.ai/premium/magic-ai Magic AI # Magic AI Application Documentation ## Table of Contents 1. Introduction * Overview of Magic AI * Key Features and Capabilities 2. Getting Started * Account Creation * Navigating the Dashboard 3. Features Overview * AI Writer * AI Image Generator * AI Code * AI Voiceover and Voice Clone * AI YouTube and AI RSS * AI Plagiarism Checker and Content Detector 4. Detailed Feature Guides * Utilizing the AI Image Generator * Leveraging the AI Writer * Engaging with AI Chat * Analyzing with AI Plagiarism Checker * Exploring AI Content Detector * AI Code Generation * Converting with AI Speech to Text * Integrating AI YouTube * Harnessing AI RSS 5. Advanced Features * Vision AI * AI File Chat * AI Tutor Dev 6. User Interface and Navigation * Dashboard Overview * Main Menu and Toolbar * Document Management 7. Collaboration and Management * Teams and Affiliates * Collaboration Features * Managing Projects and Sharing 8. Customization and Preferences * Setting Up Workflows * Custom Templates and Preferences 9. Support and Resources * FAQ Section * Tutorial Videos * Contacting Support 10. Security and Privacy * Data Handling and Privacy Policy * Compliance and Security Measures 11. Appendix and Additional Resources * Glossary of Terms * External Resources and Links 12. Conclusion and Future Directions * Summary * Upcoming Features *** ## 1. Introduction ### Overview of Magic AI Magic AI is a versatile platform designed to enhance productivity and creativity across various domains including content creation, code generation, and task automation. By leveraging advanced artificial intelligence, Magic AI provides users with tools to generate text, images, code, and much more, streamlining workflows and enabling innovation. ### Key Features and Capabilities * AI Writer: Automates content creation, offering templates and custom content generation. * AI Image Generator: Creates custom images based on user prompts. * AI Code: Generates code snippets to aid development projects. * AI Voiceover and Voice Clone: Converts text to speech and creates voice clones. * AI YouTube and AI RSS: Transforms video content into blog posts and generates content from RSS feeds. * AI Plagiarism Checker and Content Detector: Ensures content originality and detects AI-generated text. ## 2. Getting Started This section guides new users through creating an account, setting up their profile, and provides an introduction to navigating the Magic AI dashboard for the first time. \[Continue with each section as outlined, providing detailed explanations, tips, and visual aids where necessary to ensure users can easily understand and engage with Magic AI's vast functionalities.] ## 3. Features Overview ### AI Writer The AI Writer is a powerful tool designed to automate and enhance content creation. It offers a variety of templates for different content needs, including blogs, advertising copy, and more. Users can start with a template or input custom prompts to generate unique content tailored to their needs. ### AI Image Generator With the AI Image Generator, users can create stunning, custom images. By entering prompts and selecting from various art styles and resolutions, users have the power to bring their visual ideas to life, whether for professional projects or personal creative endeavors. ### AI Code The AI Code feature accelerates the development process by generating code snippets based on user requirements. It supports a wide range of programming languages and use cases, from web development to data analysis, making it an invaluable tool for developers of all skill levels. ### AI Voiceover and Voice Clone Magic AI's AI Voiceover service transforms text into natural-sounding speech, offering a selection of voices and pacing options. The AI Voice Clone feature goes a step further, allowing users to create a digital clone of their own voice for personalized audio content. ### AI YouTube and AI RSS These features allow users to extend the reach of their content. AI YouTube converts video content into engaging blog posts, while AI RSS generates fresh content from selected RSS feeds, keeping websites or blogs updated with the latest information. ### AI Plagiarism Checker and Content Detector To ensure the integrity and originality of content, the AI Plagiarism Checker scans text against a vast online database. The AI Content Detector analyzes writing to determine if it's human or AI-generated, aiding in maintaining authenticity. ## 4. Detailed Feature Guides ### Utilizing the AI Image Generator 1. Getting Started: Navigate to the AI Image Generator section from the dashboard. 2. Creating Your Image: Enter a prompt describing the image you want to create. Choose the art style and resolution. 3. Customization Options: Explore advanced settings to fine-tune your image's appearance. 4. Finalizing and Downloading: Preview your image, make any necessary adjustments, and download the final product. ### Leveraging the AI Writer 1. Selecting a Template: Choose from a wide array of content templates or opt for a custom prompt. 2. Customizing Your Content: Input the required information, such as tone, style, and specific details you want included. 3. Generating Content: Click the generate button and receive your tailored content within seconds. 4. Editing and Refinement: Use the Advanced Editor for further refinements, ensuring your content is polished and ready for publication. \[Continue in a similar detailed manner for each feature, providing step-by-step guides that empower users to fully utilize the capabilities of Magic AI.] ## 5. Advanced Features ### Vision AI Upload images to receive comprehensive descriptive analyses, extracting deeper insights and contextual understanding from visual content. ### AI File Chat Interact with uploaded documents in a conversational manner to summarize content, extract key information, or analyze data. ### AI Tutor Dev A feature designed for developers to enhance their coding skills through interactive learning sessions and personalized coding challenges. ## 6. User Interface and Navigation ### Dashboard Overview The dashboard provides a central hub for accessing all Magic AI features. It offers a quick overview of your activity, including remaining credits for text and image generation, recent projects, and easy access to all tools. ### Main Menu and Toolbar The main menu on the left-hand side allows for easy navigation between different functions like the AI Writer, AI Image Generator, and AI Code. The toolbar at the top provides quick actions such as creating new documents, accessing settings, and managing your account. ### Document Management Organize, search, and manage your documents efficiently. Features include the ability to sort by type, date, or other parameters, and the option to create folders for better organization. ## 7. Collaboration and Management Magic AI supports collaboration through features like Teams and Affiliates, allowing users to work together on projects, share documents, and manage team activities seamlessly. ## 8. Customization and Preferences Users can tailor their Magic AI experience by setting up custom content templates, defining preferred workflows, and adjusting application settings to match their individual needs. ## 9. Support and Resources ### FAQ Section A comprehensive FAQ section addresses common queries, helping users troubleshoot issues and understand the platform's capabilities. ### Tutorial Videos Embedded tutorial videos guide users through common tasks and features, enhancing understanding through visual learning. ### Contacting Support Details on how to contact the support team for personalized assistance, including email, live chat, and community forums. ## 10. Security and Privacy Magic AI is committed to user privacy and data security, adhering to global standards and regulations. This section outlines our data handling practices, privacy policy, and the security measures in place to protect user information. ## 11. Appendix and Additional Resources This section includes a glossary of terms used throughout the documentation, links to external resources, and additional materials that users may find helpful in maximizing their use of Magic AI. ## 12. Conclusion and Future Directions Magic AI is continually evolving, with new features and enhancements based on user feedback. This section summarizes the documentation, highlights the commitment to innovation, and provides a glimpse into the future of Magic AI. # MultiBot Agents Source: https://support.myapps.ai/premium/multibot-agent-builder MultiBot Agents # General Overview #### Introduction MultiBot Agents provide a **flexible framework** for creating and managing custom AI assistants powered by various model providers. This feature allows users to build **specialized AI agents** with unique configurations, capabilities, and integrations—without requiring extensive coding knowledge. MultiBot Agents function similarly to OpenAI’s Assistants API and ChatGPT’s GPTs but offer **broader model support** and a **no-code** implementation, making it easy to deploy AI-driven solutions for different tasks. This guide provides a **comprehensive walkthrough** of setting up, configuring, and utilizing MultiBot Agents effectively. *** ### Getting Started with MultiBot Agents #### Creating a New Agent To create a new agent: 1. **Open the MultiBot Agents Panel** * Navigate to the **"Agents"** section from the endpoint menu. * Click on the **Agent Builder** in the Side Panel. 2. **Fill Out Agent Details** * **Avatar:** Upload a custom avatar for personalization. * **Name:** Choose a distinctive name for your agent. * **Description:** (Optional) Add a brief summary of your agent’s purpose. * **Instructions:** Define system instructions to shape the agent’s behavior. * **Model:** Select from available AI providers and models. 3. **Select an Existing Agent** * Agents can be selected from the **top dropdown menu** in the Side Panel. * You can also mention an agent in the chat using **"@"** in the input field. *** ### Model Configuration The **Model Parameters Interface** allows fine-tuning of your agent's responses: * **Temperature (0-1):** Adjust response creativity. * **Max Context Tokens:** Set the memory limit for the conversation. * **Max Output Tokens:** Define the maximum response length. * **Provider-Specific Settings:** Customize parameters based on the chosen AI model. *** ### Agent Capabilities #### **1. Code Interpreter** When enabled, the **Code Interpreter** allows your agent to: * Execute code in multiple programming languages, including:**Python, JavaScript, TypeScript, Go, C, C++, Java, PHP, Rust, and Fortran.** * Process files securely through the **MultiBot Code Interpreter API**. * Run code without requiring local setup, configuration, or sandbox deployment. * Handle file uploads and downloads seamlessly. *** #### **2. File Search** The **File Search** feature enables: * **Retrieval-Augmented Generation (RAG):** AI can retrieve relevant information from uploaded documents. * **Semantic Search:** AI understands document content contextually. * **Context-Aware Responses:** AI can reference uploaded files in conversations. * **File Attachment Support:** Attach documents at both the **agent** and **chat thread** levels. *** ### Tools & Integrations MultiBot Agents can be enhanced with various built-in tools: * **DALL-E 3:** Generate images from text descriptions. * **Tavily Search:** Advanced search API with diverse data sources. * **Calculator:** Perform mathematical calculations. * **Google Search:** Retrieve real-time web search results. * **Stable Diffusion:** Generate images from text prompts. * **Azure AI Search:** Retrieve structured information. * **Traversaal:** A robust search API for LLM Agents. * **Wolfram Alpha:** Perform complex computations and mathematical reasoning. *** ### Actions & Custom API Integrations With **Actions**, you can extend your agent’s functionality by adding **custom tools** from OpenAPI specifications. #### **Adding Custom Actions** 1. Click the **"Create Action"** button in the Agent Builder. 2. Enter the **OpenAPI spec URL** to generate a new tool. 3. Save and assign the action to your agent. > **Note:** Individual domains can be **whitelisted** for agent actions. *** ### File Management MultiBot Agents support **three types of file uploads**: 1. **Image Uploads:** For AI-based visual content processing. 2. **File Search Uploads:** Upload documents for **RAG** capabilities. 3. **Code Interpreter Uploads:** Upload files for **code execution**. Files can be attached either: ✅ **Directly to the agent configuration** ✅ **Within individual chat threads** *** ### Best Practices To ensure the best experience with MultiBot Agents: ✅ **Provide clear, specific instructions** for your agent. ✅ **Carefully select the necessary tools** for your use case. ✅ **Organize files properly** across the three upload categories. ✅ **Review sharing permissions** before making an agent public. ✅ **Test your agent thoroughly** before deploying it. *** ### Recap: How to Create & Use MultiBot Agents 1. **Select "Agents"** from the endpoint dropdown menu. 2. **Open the Agent Builder panel.** 3. **Fill in the required details** (name, avatar, model, instructions). 4. **Configure capabilities** (Code Interpreter, File Search, Tools). 5. **Add necessary tools and files.** 6. **Set sharing permissions** (if needed). 7. **Create & start using your agent!** 🎉 *** ### Conclusion MultiBot Agents provide a **powerful, no-code** solution for building AI-driven assistants with **customizable capabilities**. Whether you're creating **chatbots, automation tools, or research assistants**, MultiBot Agents offer **seamless integration** with your preferred AI models and services. By following this guide, you can **quickly set up, deploy, and scale** AI agents tailored to your specific needs. *** # Multibot Intro Source: https://support.myapps.ai/premium/multibot-intro Multibot Intro ## Overview Multibot is an AI-driven platform that allows users to engage with multiple chatbot personas simultaneously. It's designed to provide a seamless interaction experience for various purposes, including customer support, personal assistance, and more. ## User Interface Components ### Sidebar Navigation * New Chat (+): Initiates a new chat session with a selected bot persona. * Chat List: Shows the active chat sessions with different bot personas. Users can switch between chats by clicking on them. ### Bot Selection Panel * Bot Categories: Users can select a bot from a range of categories, such as `OpenAI`, `Mistral`, `OpenRouter`, `Google`, `Bing`, and `Plugins`, to initiate or switch to a chat session with the chosen bot. * Active Bot Indicator (Google logo): Highlights which bot is currently active in the chat window. * Close Sidebar: Collapses the bot selection panel for an expanded view of the chat area. ### Chat Window * Chat Input Field: Where users type their messages or queries to the selected bot persona. * Send Button (paper airplane icon): Sends the typed message to the bot. * Chat Output Area: Displays the conversation with the bot, including both user input and bot responses. * Image Insertion: Where users can input an image that the AI will read and respond to. * To do this navigate to the paperclip icon at the left side of the prompt bar. This is where you'll insert your image, however, some images may be too big. * Head over to [https://squoosh.app](https://squoosh.app) where you can make your image smaller and ready to be read by the AI. ![](https://docs.myapps.ai/storage/page_media/0cac99b6-2b42-4210-9be5-482acfd4d182.png) ### Bot Settings and Customization Panel * Model Dropdown: Selects the AI model for the bot persona. * Temperature Control: Adjusts the randomness of the bot's responses. * Top P / Top K Sliders: Controls the diversity of the bot's responses. * Max Output Tokens: Sets the maximum length of the bot's response. * Custom Name Field: Assigns a custom name to the bot session for easy identification. * Prompt Prefix Field: Sets a custom prompt or instruction for the bot to follow during the chat session. * Save As Preset: Saves the current settings as a preset for future use. * Import Button: Allows importing settings from a file. ### Conclusion Multibot is a versatile tool that amalgamates the functionalities of diverse AI models into a unified chat interface, empowering users to conduct multiple conversations and manage interactions efficiently. Its user-friendly interface ensures a streamlined experience while offering advanced settings for customized interactions. # Navigating Vendors Source: https://support.myapps.ai/premium/multibot-vendors Navigating Vendors in Multibot Navigating through vendors and selecting models in a platform designed for integrating AI functionalities into applications or services can be a crucial step in tailoring your project's capabilities to specific needs. This process typically involves exploring a diverse ecosystem of AI models provided by various vendors, each offering unique strengths and functionalities. Below is a guide to effectively navigate through vendors and choose models, using the example of selecting the latest version of "Claude" from OpenRouter. image.png ### Navigating Through Vendors * Access the Vendor Dropdown Menu: Start by locating the dropdown menu at the top left of the platform's interface. This menu is your gateway to the extensive list of AI model vendors available on the platform, such as OpenRouter, Groq, Google, and others. ![](https://docs.myapps.ai/storage/page_media/a6d80501-0f8d-4bb1-aa43-9fd4643100c5.png) * Select Your Desired Vendor: Click on the dropdown menu to explore the list of vendors. Each vendor specializes in different types of AI models and services. For instance, if you're interested in cutting-edge language models, you might choose a vendor known for their advancements in natural language processing. * Understand Vendor Specializations: Before making a selection, it's helpful to have a basic understanding of what each vendor is known for. This information can usually be found on the platform or by doing a quick web search. Selecting the right vendor is crucial for finding the model that best fits your project's requirements. Choosing Models from a Vendor * Navigate to the Model Search Bar: Once you've selected a vendor, move to the search bar located next to the vendor dropdown menu. This search bar allows you to explore the specific models offered by the selected vendor. * Search for Specific Models: If you already have a model in mind, such as the latest version of "Claude" from OpenRouter, you can directly type the model's name into the search bar. Alternatively, you can browse through the list of models if you're exploring options or looking for inspiration. ![](https://docs.myapps.ai/storage/page_media/13e80c2d-103d-4ed4-ad75-fa658ae22afa.png) * Review Model Details: Upon finding the model you're interested in, take the time to review its details, capabilities, and any version notes. This step is crucial for ensuring the model meets your technical and functional requirements. * Select the Model: After reviewing the model's details and confirming it aligns with your project's needs, select the model for integration. Depending on the platform, this may involve clicking a "Select" button, downloading the model, or following specific integration instructions. ### Tips for Effective Model Selection * Consider Compatibility: Ensure the model you select is compatible with your project's existing architecture and technology stack. * Review Documentation: Take advantage of any available documentation or user guides provided for the model. This can offer valuable insights into integration processes and best practices. * Explore Community Feedback: Look for forums, reviews, or community feedback about the model and vendor. Insights from other users can highlight potential strengths and limitations. Navigating through vendors and choosing the right models is a pivotal process in leveraging AI to enhance your projects. By following these steps and tips, you can make informed decisions that align with your project's goals and technical requirements, ensuring a successful integration of AI functionalities. # o1 Beta Source: https://support.myapps.ai/premium/o1-beta Multibot with access to the o1 preview Open AI model # o1 Beta Documentation ## Table of Contents 1. [Overview](#overview) 2. [Interface Layout](#interface-layout) 3. [Left Sidebar Menu](#left-sidebar-menu) 4. [AI Provider, Model Selection, and Presets Configuration](#ai-provider-model-selection-and-presets-configuration) 5. [Right Sidebar Menu](#right-sidebar-menu) 6. [Chat Interface](#chat-interface) 7. [OpenAI o1 Series Models](#openai-o1-series-models) 8. [Advanced Features](#advanced-features) 9. [Troubleshooting](#troubleshooting) 10. [FAQ](#faq) ## 1. Overview o1 Beta is a powerful multibot platform with access to the cutting-edge o1 Beta LLM (Language Learning Model). This application allows users to interact with various AI models, manage chats, and customize their AI experience. ![Overview](https://img.mytsi.org/i/Yfwi767.png) ## 2. Interface Layout The o1 Beta interface is divided into three main sections: * Left Sidebar: For navigation and chat management * Central Area: For chat interactions * Right Sidebar: For additional features and settings ## 3. Left Sidebar Menu ![Left Sidebar Menu](https://img.mytsi.org/i/KY0A768.png) The left sidebar provides quick access to essential features: 1. **Chat Management** * Create a new chat * View chat history 2. **Bookmarks** * Access saved conversations or important information 3. **Account Information** * View and manage your account details ## 4. AI Provider, Model Selection, and Presets Configuration ![AI Provider and Model Selection](https://img.mytsi.org/i/OjWU769.png) ### 4.1 AI Providers Selection Choose from a variety of AI providers to power your conversations. ![AI Providers](https://img.mytsi.org/i/1JVg771.png) ### 4.2 Model Selection Access a dropdown menu to view and select from all available models, including the latest Open AI's o1 Preview model. ![Model Selection](https://img.mytsi.org/i/OFAv772.png) ### 4.3 Presets Customize and save your preferred settings for quick access. ![Presets](https://img.mytsi.org/i/gKBM773.png) Access your stored presets for efficient workflow. ![Stored Presets](https://img.mytsi.org/i/FZEb774.png) ## 5. Right Sidebar Menu ![Right Sidebar Menu](https://img.mytsi.org/i/OGW9770.png) The right sidebar offers additional features to enhance your o1 Beta experience: ### 5.1 Setting up Prompts Customize prompts for your AI model to guide conversations effectively. ![Setting up Prompts](https://img.mytsi.org/i/nlgQ775.png) ### 5.2 Managing Prompts Organize and edit your prompts for optimal use. ![Managing Prompts](https://img.mytsi.org/i/8fke776.png) ### 5.3 Attach Files Easily attach files to your conversations for context or analysis. ![Attach Files](https://img.mytsi.org/i/hBTC777.png) ### 5.4 File Management Manage your uploaded files efficiently. ![File Management](https://img.mytsi.org/i/AjxB777.png) ### 5.5 Manage Bookmarks Organize and access your bookmarked conversations or information. ![Manage Bookmarks](https://img.mytsi.org/i/4VWd779.png) ## 6. Chat Interface The central area of the o1 Beta interface is dedicated to your chat interactions with the AI. Here, you can: * Type your messages * View AI responses * Scroll through conversation history ## 7. OpenAI o1 Series Models The o1 Beta platform now includes access to OpenAI's groundbreaking o1 series models, which represent a significant advancement in AI reasoning capabilities. ### 7.1 Overview of o1 Models OpenAI o1 series models are new large language models trained with reinforcement learning to perform complex reasoning. These models excel in: * Scientific reasoning * Competitive programming (89th percentile on Codeforces) * Advanced mathematics (top 500 students in USA Math Olympiad qualifier) * PhD-level accuracy on physics, biology, and chemistry problems ### 7.2 Available o1 Models Two reasoning models are available in the API: 1. **o1-preview**: An early preview designed to reason about hard problems using broad general knowledge. 2. **o1-mini**: A faster and cheaper version, particularly adept at coding, math, and science tasks where extensive general knowledge isn't required. ### 7.3 Key Features of o1 Models * **Internal Reasoning**: o1 models think before they answer, producing a long internal chain of thought. * **Reasoning Tokens**: These models use reasoning tokens to break down their understanding and consider multiple approaches. * **Large Context Window**: Both models offer a context window of 128,000 tokens. ### 7.4 Using o1 Models in o1 Beta To use o1 models in your o1 Beta application: 1. Select the desired o1 model from the model dropdown menu. 2. Craft your prompt or question. 3. Submit and wait for the response (response times may vary based on complexity). ### 7.5 Best Practices for o1 Models * Keep prompts simple and direct. * Avoid chain-of-thought prompts or instructions to "think step by step." * Use delimiters for clarity in complex prompts. * Limit additional context in retrieval-augmented generation (RAG). ### 7.6 Limitations During Beta * Text-only input (no image support) * Limited message types (user and assistant only) * No streaming support * No tool or function calling support * Fixed parameters for temperature, top\_p, and n ### 7.7 Managing Costs and Context * Use the `max_completion_tokens` parameter to control total tokens generated. * Reserve at least 25,000 tokens for reasoning and outputs when starting. * Monitor the usage object in responses for token consumption details. ## 8. Advanced Features \[Content remains the same as in the previous version, with the addition of:] * **Complex Reasoning**: Leverage o1 models for tasks requiring deep analysis and multi-step problem-solving. * **Code Generation and Refactoring**: Utilize o1 models for advanced coding tasks and code improvements. ## 9. FAQ **Q: What makes o1 Beta unique?** A: o1 Beta offers access to cutting-edge AI models, including the o1 Preview model, with a highly customizable interface. **Q: Can I switch between AI providers mid-conversation?** A: Yes, you can switch providers, but note that context may not carry over between different AI models. **Q: Is there a limit to how many conversations I can have?** A: Limits may vary based on your account type. Check your account details for specific information. **Q: Can I export my conversations?** A: Yes, o1 Beta allows you to export your chat history in various formats. **Q: How do o1 models differ from GPT-4o models?** A: o1 models excel in complex reasoning tasks but may have longer response times. GPT-4o models are better for applications requiring image inputs, function calling, or consistently fast responses. **Q: Can I see the reasoning process of o1 models?** A: The internal reasoning process (reasoning tokens) is not visible in the API output, but it contributes to the final response quality. **Q: How can I optimize my prompts for o1 models?** A: Keep prompts simple and direct, avoid instructing the model to explain its reasoning, and use clear delimiters for complex inputs. *** # Premium Assistants Source: https://support.myapps.ai/premium/premium-assistants Premium Assistants # Premium & Assistants ### Introduction to AI Tutor Premium AI Tutor Premium stands out as an advanced platform designed to enhance user interaction and efficiency through the use of sophisticated AI models and personalized AI assistants. This premium version builds upon the features available in MultiBot and Plus, offering more tailored functionalities to users seeking a higher degree of customization and control over their AI interactions. With AI Tutor Premium, users can program the AI with specific prompts, choose from a variety of models for different tasks, and notably, create and utilize AI assistants for optimized chat responses and task management. ### Key Features of AI Tutor Premium * Customizable AI Settings: Users have the flexibility to program the AI with specific prompts and select the most suitable model for their needs, ensuring tailored responses and functionalities. * AI Assistants: A standout feature of AI Tutor Premium is the introduction of AI assistants. These assistants can be programmed with descriptions, prompts, images, and even specific files, acting as personalized AI entities to support users in their tasks. * Comprehensive Sidebar Tools: The left sidebar serves as the control panel for accessing various features, including chats, prompts, tools, files, and collections, providing a centralized location for managing AI interactions. * Optimized Chat Responses: By leveraging AI assistants within prompts, users can significantly enhance the quality and efficiency of chat responses, simulating the work of multiple people within a single interaction. ### How to Use AI Tutor Premium Access AI Settings: Click on the icon at the top right to enter the AI settings. Here, you can input your desired prompt and select the AI model that best fits your requirements. * *Please note GPT 4o can be selected to talk to images.* ![](https://docs.myapps.ai/storage/page_media/635721fa-d51c-48ac-8620-b6ee940ea654.png) * Note that there are also advanced settings that can modify the temperature of the bot (creative or factual information ). You can also set a limit on usage tokens as well. You can also choose whether or not you'd like your chats to use information from your profile context or your workspace instructions. ![](https://docs.myapps.ai/storage/page_media/ee55c899-6749-473b-8b4c-c510611d4df9.png) * When prompting you can also insert files and images to be read by the chat bot. In order for your bot to read images make sure you change your model to "gpt vision". * *Please note when inserting images that are too big errors may occur. Please use* [*https://squoosh.app/*](https://squoosh.app/)*to compress large images for chat bot use.* ### Creating AI Assistants * Create or Access AI Assistants: To access or create a new AI assistant, click on the calendar icon on the left sidebar. This area allows you to manage and program your AI assistants. ![](https://docs.myapps.ai/storage/page_media/412eccdb-51cf-4eb5-bbc6-1a7da51837ec.png) * Program Your AI Assistant: Give your AI assistant a name and detailed description. This is a short description to inform you of your Assistant. * Next input specific prompts. This is where you'll program a purpose in your assistant. What do you want it to help you with? * After that you can upload any relevant files or collections you'd like it to use. You can add these under the "files" and "collections" sections. You may also add tools and choose the model of your choosing for your assistant. * This process tailors the assistant to perform tasks or generate responses aligned with your needs. * Integrate AI Assistant in Prompts: Once your AI assistant is created and programmed, you can utilize it within your prompts to optimize chat responses. This integration enhances the AI's ability to provide relevant, accurate, and efficient interactions. ### Tips for Maximizing the Use of AI Tutor Premium * Be Detailed in Programming: The more detailed and specific you are when programming your AI and assistants, the more accurate and helpful their responses and actions will be. * Leverage AI Assistants for Specialized Tasks: Create different AI assistants for various tasks or topics. This specialization allows for more focused and expert responses in specific areas. * Use Insert File Tool: Input any file and ask you chatbot information based on the uploaded file. Your assistant may also help you do anything you need their help with concerning said file or information. * Regularly Update Your AI Assistants: As your needs evolve, make sure to update your AI assistants' programming with new information, prompts, and files to keep them effective and relevant. * Experiment with Different Models: Don't hesitate to try out different AI models for various tasks. Each model may offer unique strengths and capabilities that could benefit your specific use case. ### Conclusion AI Tutor Premium offers a powerful suite of features for users looking to harness the capabilities of AI for personalized and efficient interactions. Through customizable settings, the innovative use of AI assistants, and a comprehensive set of tools, AI Tutor Premium enables users to optimize their chat responses and automate tasks, effectively enhancing productivity and engagement. Whether for business, education, or personal use, AI Tutor Premium provides the tools and flexibility needed to make the most out of AI technology. # Prompt Gallery Source: https://support.myapps.ai/premium/prompt-gallery Prompt Gallery ## Introduction The Prompt Gallery is a curated collection of AI prompts designed to streamline and inspire the creation of content across various domains. ## Home Page Upon accessing the Prompt Gallery, users are greeted with a colorful grid layout. ### Grid Layout * Prompt Categories: Each cell in the grid represents a different prompt category such as Writing, Marketing, Backend Development, etc. * Interactive Cells: Clicking on a cell expands to reveal a set of specialized prompts related to the chosen category. ## Prompt Details When a category cell is selected, users can view detailed prompts: ### Expanded View * Prompt Title: The name of the prompt for easy identification. * Prompt Description: A brief description of what the prompt entails. * Copy Button: Allows users to copy the prompt to the clipboard. * Show Button: Reveals additional details or examples related to the prompt. ## Categories Page Selecting the 'Categories' tab on the navigation bar directs users to a comprehensive list of available prompt categories. ### Categories Layout * List of Categories: Organized in alphabetical order for quick browsing. * Search Functionality: Users can search for specific categories. ## Detailed Prompt View Upon selecting a specific prompt, the application presents a detailed view. ### Detailed Explanation * Full Prompt Description: An in-depth explanation of the prompt, including parameters and expected outcomes. * Copy Full Prompt: This button allows the entire prompt, including its detailed description, to be copied for use. ## User Interaction * Copy Function: Users can copy prompts to use in their projects. * Show More: Expands the prompt to show full details. ## Dashboard Users can navigate back to the main grid view by selecting the 'Dashboard' tab, which provides: ### Dashboard Overview * Quick Access: Immediate return to the interactive grid of prompts. * Recent Prompts: Displays the most recently viewed or added prompts. ## Conclusion The Prompt Gallery serves as an expansive resource for AI prompts, offering users a diverse range of templates to kickstart their AI-driven projects. With its user-friendly interface, the Prompt Gallery ensures a smooth and efficient content creation process. # Team AI Source: https://support.myapps.ai/premium/team-ai Team AI ### Introduction Step into the future of work efficiency and team dynamics with TeamAI. Assemble your ultimate team of intelligent bots, a seamless process facilitated by our advanced platform. We offer exceptional customization, empowering you to architect, provision, and activate a personalized fleet of AI agents that are fine-tuned to your specific objectives and requirements. ### Customize Your Crew * Roles & Skills: Assign specific roles and skills to your bots, ensuring a well-rounded team capable of tackling any challenge. * Information Integration: Equip your bots with any information they need. From industry-specific knowledge to company databases, they'll be ready to serve. * Unique Personalities: Give your bots distinct personalities to match your company culture or to provide a diverse set of interactive experiences. ![](https://docs.myapps.ai/storage/page_media/f328b5d8-c498-4b1b-aa35-9e9a5db0d128.png) ### Design With Freedom * Visual Customization: From avatars to UI elements, personalize the appearance of your bots and the interface from which you control them. * Interaction Styles: Set the tone for your bots interactions with users. Pick the language, the tone, and the overall style to design a communication approach that truly represents your brand. ### Security and Compliance * Data Protection: Your and your users privacy is paramount. Our bots are designed with top-tier security measures to keep data safe. * Compliance Ready: AI Teams is built to comply with industry standards and regulations, ensuring your bot team operates within legal boundaries. Get started with AI Teams today and transform your digital workforce with bots that are as unique as your business needs. Unleash potential, automate intelligently, and provide unparalleled service with your very own AI team. ### Ready to Build Your Team? # Typebot Source: https://support.myapps.ai/premium/typebot Typebot # Typebot ## Overview Typebot is a versatile chatbot creation platform designed to facilitate engaging conversations between businesses and their audience. The platform is intuitive, allowing users to craft chatbots without needing extensive programming knowledge. ## Interface and Navigation * Dashboard: Upon entering Typebot, the dashboard presents users with a clean workspace. Here, users can create a new Typebot or access existing projects. * Settings & Members: This section allows users to manage global settings and collaborate with team members. * Create a Typebot: The central feature of the interface. Users can start from scratch, use a pre-built template, or import a file to begin their Typebot journey. * Theme Customization: Users can personalize the look and feel of their Typebot. The theme settings include options for fonts, colors, and the inclusion of custom CSS for advanced styling. * Chat Flow Construction: Typebot offers a visual flow editor to design the conversation path the bot will take. This includes crafting messages, collecting user input, and setting up logic for dynamic interactions. * Publishing: Once the Typebot is configured and tested, users can publish it to make it live. The process is straightforward, with a clear button prominently displayed. ## Features and Functionalities * Templates: A gallery of templates is available to jump-start the Typebot creation process, catering to various use cases like customer support, lead generation, and surveys. * Customization Tools: Users can adjust the design of chat bubbles, user avatars, buttons, and input fields to align with their branding. * Typing Emulation: This feature simulates human typing, enhancing the chatbot's realism. * Integrations: Typebot seamlessly integrates with various services like Google Sheets, allowing for easy data management and actions based on user responses. * Plugin Support: Users can enhance their Typebots with plugins for added functionality, such as Google Tag Manager or custom APIs. ## Conclusion Typebot stands out with its user-friendly interface, comprehensive customization options, and powerful integration capabilities, making it an excellent tool for creating sophisticated chatbots tailored to specific business needs. # Video Sound FX Source: https://support.myapps.ai/premium/video-sound-fx Video Sound FX # Video Sound FX Documentation ## Table of Contents 1. [Overview](#overview) 2. [Getting Started](#getting-started) 3. [Uploading Your Video](#uploading-your-video) 4. [Generating Sound Effects](#generating-sound-effects) 5. [Best Practices](#best-practices) 6. [Troubleshooting](#troubleshooting) 7. [FAQ](#faq) ## 1. Overview Video Sound FX is an innovative tool designed to enhance your videos with custom-generated sound effects. This application uses advanced AI technology to analyze your video content and create appropriate sound effects, adding a new dimension to your visual storytelling. ## 2. Getting Started To begin using Video Sound FX, navigate to the main dashboard. Here, you'll find options to start a new project or continue working on existing ones. ## 3. Uploading Your Video The first step in using Video Sound FX is to upload your video file. ![Upload Video Interface](https://img.mytsi.org/i/d2Zz726.png) * Click on the "Upload Video" button or drag and drop your video file into the designated area * Supported video formats include MP4, AVI, MOV, and WMV * A progress bar will indicate the upload status ## 4. Generating Sound Effects Once your video is uploaded, Video Sound FX will analyze the content and generate four different sound effect options. ![Sound Effects Options](https://img.mytsi.org/i/mBHj725.png) * After processing, you'll be presented with four unique sound effect tracks * Listen to each option by clicking the play button next to it * Choose the sound effect that best fits your video by selecting the corresponding option * You can preview how the selected sound effect works with your video before finalizing ## 5. Best Practices To get the most out of Video Sound FX, consider the following best practices: ### 5.1 Video Simplicity * Keep your video content simple and focused * Complex scenes with multiple actions may result in less accurate sound effects ### 5.2 Optimal Video Length * For best results, use videos that are under 10 seconds long * Shorter videos allow the AI to focus on specific actions and generate more accurate sound effects ### 5.3 Clear Action Sequences * Ensure that the main actions in your video are clearly visible * This helps the AI accurately interpret the scene and generate appropriate sound effects ### 5.4 Good Video Quality * Use high-quality video footage for better analysis * Clear, well-lit scenes provide better input for the AI to work with ## 6. Troubleshooting If you encounter any issues while using Video Sound FX, try the following: * Ensure your video file is in a supported format * Check your internet connection for smooth uploading and processing * Clear your browser cache and try again * If problems persist, contact our support team ## 7. FAQ **Q: What video formats are supported?** A: Video Sound FX supports MP4, AVI, MOV, and WMV formats. **Q: Is there a limit to the video length?** A: While there's no strict limit, we recommend using videos under 10 seconds for optimal results. **Q: Can I use copyrighted music in my videos?** A: Video Sound FX generates original sound effects. For any background music, ensure you have the necessary rights or licenses. **Q: How long does it take to generate sound effects?** A: Processing time varies depending on video length and complexity, but typically takes a few minutes. **Q: Can I edit the generated sound effects?** A: Currently, Video Sound FX doesn't offer editing capabilities. You can choose from the four generated options. *** # Workspace Instructions Source: https://support.myapps.ai/premium/workspace-instructions Workspace Instructions & Profile Context # Workspace Instructions & Profile Context ### Exploring Additional Features of AI Tutor Premium: Workspace and Profile Instructions AI Tutor Premium goes beyond basic AI interactions by offering users the ability to personalize their experience through Workspace and Profile Instructions. These features allow for a deeper level of customization, ensuring that the AI's responses and functionalities are closely aligned with the user's or company's specific needs and contexts. By setting up detailed instructions in both the user profile and workspace, users can significantly enhance the relevance and accuracy of AI-generated responses. ### Accessing and Utilizing Profile Instructions * Access Profile Instructions: To begin personalizing your AI interactions, click on the user icon located at the bottom left of the sidebar. This section is dedicated to setting up your user profile within AI Tutor Premium. ![](https://docs.myapps.ai/storage/page_media/b5a4023d-e68a-44ec-b118-c302d90d837c.png) * Add Personal or Company Information: In the profile instructions area, you have the opportunity to add information about yourself or the company you represent. This could include your role, areas of expertise, or any other relevant details. * Customize Response Preferences: Crucially, within the description, you can specify how you'd like your AI-generated responses to be formulated. Whether you prefer concise answers, detailed explanations, or a particular tone, this is where you can guide the AI's output style. ### Setting Up Workspace Instructions * Navigate to Workspace Instructions: Workspace instructions can be found under the gear icon on your page. This section is designed for you to provide context about the subjects or projects you'll be focusing on within your workspace. ![](https://docs.myapps.ai/storage/page_media/a435aa41-bb51-4a8b-828e-99fa57cf0f0f.png) * Provide Contextual Information: Here, you can add specific context about the topics you'll be chatting about or the tasks you'll be using the AI to assist with. This information helps the AI tailor its responses and functionalities to your workspace's unique requirements. * Influence AI Responses: By setting up workspace instructions, you ensure that the AI's responses are informed by the specific context of your projects or discussions, leading to more relevant and useful interactions. ![](https://docs.myapps.ai/storage/page_media/186f4b2e-3ac9-43d5-b48a-a6293c7b553c.png) ### Deactivating Personalized Instructions * Adjust Settings in Advanced Options: If, for any reason, you decide to deactivate the personalized instructions you've set up, you can do so in the advanced settings. This option is located in the models tab, as previously mentioned. Deactivating these instructions will revert the AI's responses to more general outputs, not tailored by your profile or workspace settings. ### Tips for Effective Use of Profile and Workspace Instructions * Be Detailed and Specific: The more detailed and specific you are in your profile and workspace instructions, the more accurately the AI can align its responses and functionalities with your needs. * Regularly Update Instructions: As your projects evolve or your company's focus shifts, make sure to update your instructions to keep the AI's responses relevant and aligned with current objectives. * Balance Personalization and Flexibility: While personalizing your AI interactions can significantly enhance the tool's usefulness, also consider maintaining some level of flexibility in the AI's responses for unexpected queries or tasks. ### Conclusion The Profile and Workspace Instructions features of AI Tutor Premium provide users with powerful tools for customizing AI interactions to fit their specific needs and contexts. By carefully setting up and maintaining these instructions, users can ensure that AI-generated responses are not only accurate and relevant but also reflective of their personal or company identity. Whether for enhancing productivity, facilitating learning, or supporting creative projects, these customization options make AI Tutor Premium a versatile and adaptable platform for a wide range of applications. # Quickstart Source: https://support.myapps.ai/quickstart From sign-up to your first generation in about five minutes This walks you through your first generation, then shows you where it went. ## 1. Create your account Go to [beta.pixio.myapps.ai](https://beta.pixio.myapps.ai) and sign up with email or a social provider. New accounts start with **50 free credits that never expire**. 50 credits is real working budget if you start with an image model — many cost only a handful of credits per run. Video costs considerably more, so learn the controls on images first. ## 2. Open Generate Open **Generate** from the top navigation. You’ll see three panels: Choose your model and fill in its inputs. The current result, at full size. Everything you've generated, newest first. ## 3. Pick a model In the left panel, browse or search the model list. Models are grouped by the family that makes them and labelled with what they do — text-to-image, image-to-video, and so on — plus the credit cost per run. Start with a **text-to-image** model. Tap the star on any model to favorite it so it stays at the top of your list. Some models are limited to certain plans. Anything your plan can't run is marked in the list, and [upgrading](/credits) unlocks it. ## 4. Write your prompt and generate Fill in the prompt and any other inputs the model exposes — dimensions, aspect ratio, style, seed. Every input has a tooltip explaining what it does. Then hit generate. The run appears immediately in the right column as **processing**, and the credit cost is shown up front. The job is queued and running. You can keep working — leaving the page doesn't cancel it. The result appears in the preview and in your history. If a run fails, you aren't charged for it. Try again or adjust your inputs. ## 5. Do something with the result Hover a finished item in the right column and you get **chained actions** — one click sends that output straight into the next model, no re-uploading: * **Regenerate** with the same prompt and inputs * **Create Video** from a still image * **Next Frame** to continue a sequence * **Upscale** or **Image Upscale** for more resolution * **Remove BG** to cut out the background * **Add Lipsync** or **Generate Voice** * **Create 3D Model** from an image * **Photo Restoration** or **Edit Image** Video results get their own set — **Upscale Video**, **Caption Video**, **Lipsync Video**, **Edit Video**, **Remove BG**. Chaining is the fastest way to work: generate a cheap image first, then push the one you like into an expensive video model. Much cheaper than generating video after video. ## 6. Find it later | Where | What's there | | ------------------ | ------------------------------------------------------ | | **My Assets** | Every generation and every file you've uploaded. | | **Prompt History** | Every prompt you've used, by category, ready to reuse. | | **Boards** | Generations you've organized into a project. | You can also **download** any result, or **publish** it to the public gallery — and unpublish it later. ## What next Uploads, favorites, plan gating, and how credits are charged. What things cost and what each plan includes. How to choose between 500+ models. Save a multi-step chain and run it repeatedly. # Status Source: https://support.myapps.ai/status Live service availability, uptime, and incident updates The [status page](https://beta.pixio.myapps.ai/status) shows live service availability, uptime, and incident updates. Check it first when something isn't working — it usually answers the question faster than a support message. ## Service groups Status is reported per area, so you can tell whether the problem is Pixio-wide or specific to what you're doing: * **Image generation** * **Video generation** * **Audio generation** * **Workflow automations** A single group degrading is common — video can be struggling while images are fine. ## What you'll see | | | | ------------------------- | ---------------------------------------------------------------------------------------- | | **Active incidents** | What's degraded right now | | **Recovered in last 72h** | Recently resolved — useful if a run failed an hour ago | | **Per-model status** | [Model status](https://beta.pixio.myapps.ai/status/models) drills into individual models | | **History** | Past incidents over time | Two messages worth understanding: * *"A large share of requests is failing right now"* — real disruption; wait rather than burning credits on retries * *"No meaningful traffic in the last 24 hours"* — not enough volume to judge, not an outage ## When a generation fails If the relevant group shows an incident, that's your answer — wait it out. **Failed runs aren't charged.** [Model status](https://beta.pixio.myapps.ai/status/models) can show one model failing while its group looks healthy. Same capability, different family — usually the fastest route to an unblocked result. If status is clean and the failure repeats, it's worth reporting. Include what you ran and when. A model can be degraded upstream while everything else works normally. That's why per-model status exists — check it before concluding your inputs are wrong. ## Related * [Credits & Plans](/credits) — failed generations aren't charged * [FAQ & Support](/faq/support) — getting help # AI Console Source: https://support.myapps.ai/ultra/ai-console-ultra AI-powered full-stack web development directly in your browser. # General Overview #### Introduction **AI Console** is an advanced, AI-powered development environment that enables **full-stack web development directly in your browser**. It supports **multiple LLMs** and offers an **extensible architecture** for seamless integration of additional models. With powerful features like **image attachments in prompts, integrated terminals, version control, and Docker support**, AI Console provides a **streamlined coding experience** for developers, researchers, and AI enthusiasts. This guide walks you through the **core features, integrations, and upcoming enhancements** in AI Console. *** ### Key Features #### **1. AI-Powered Full-Stack Development** * Develop and **execute code directly in your browser** using AI-assisted tools. * Supports **multiple programming languages** and frameworks. #### **2. Multi-LLM Support** * AI Console integrates with **various large language models (LLMs)**. * Extensible architecture allows **easy integration of new models**. #### **3. Image Attachments for Contextual Prompts** * Attach **images to prompts** to provide better context to LLMs. * Enhances AI’s **understanding of visual elements** in development. #### **4. Integrated Terminal for LLM Command Execution** * View real-time **command execution output** within the console. * Debug and analyze AI-generated commands **without leaving the interface**. #### **5. Code Version Control & Revert** * Easily **revert code to earlier versions** for debugging. * Enables **quick iteration and error recovery**. #### **6. Project Portability & ZIP Downloads** * Download entire projects as **ZIP files** for easy sharing and portability. * Ensures **seamless migration** between different development environments. #### **7. Docker Integration for Hassle-Free Setup** * AI Console is **containerized with Docker** for easy installation. * Enables **quick deployment** on local machines or cloud environments. *** ### Project Management & Community Contributions **AI Console is a community-driven project!** While the core team organizes development efforts, contributions from the community help shape its future. If you’re interested in contributing, check out the **Project Management Guide** to get started. *** ### Recent & Upcoming Integrations The AI Console team is actively working on **new integrations and enhancements** to improve functionality. Below is a list of requested and implemented features: #### ✅ **LLM & API Integrations** * **OpenRouter Integration** * **Gemini Integration** * **Mistral API Integration** * **Cohere Integration** * **DeepSeek API Integration** * **xAI Grok Beta Integration** * **HuggingFace Integration** * **AWS Bedrock Integration** * **Together Integration** * **Perplexity Integration** #### ✅ **AI Development Enhancements** * **Autogenerate Ollama models** from downloaded files * **Filter models by provider** * **Dynamic model max token length** * **Better prompt enhancing** * **Prompt caching** for improved efficiency * **Attach images to prompts** for better context * **PromptLibrary** for multiple variations of prompts #### ✅ **Code & Project Management** * **Ability to sync files (one-way sync) to local folder** * **Load local projects into the app** * **Git Clone button** for easy repository access * **Git Import from URL** * **Detect `package.json` and auto-install dependencies** * **Selection tool to target changes visually** #### ✅ **Debugging & Terminal Enhancements** * **Bolt terminal** to view LLM command execution * **Streaming of code output** * **Detect terminal errors and ask AI to fix them** * **Detect preview errors and ask AI to fix them** #### ✅ **Deployment & Cloud Features** * **Containerize the application with Docker** for easy installation * **Publish projects directly to GitHub** * **Ability to enter API keys in the UI** * **Add Starter Template Options** for quick project setup #### ✅ **User Experience & Accessibility** * **Mobile-friendly UI improvements** * **Chat history backup & restore functionality** *** ### Best Practices for Using AI Console To maximize your efficiency with **AI Console**, follow these best practices: ✅ **Use structured prompts** to guide AI-generated code.\ ✅ **Attach images** where necessary for better contextual understanding.\ ✅ **Regularly revert and compare code versions** for debugging.\ ✅ **Utilize the integrated terminal** to view AI-run commands in real time.\ ✅ **Leverage Git integration** for version control and collaboration.\ ✅ **Deploy with Docker** for a stable and consistent development environment. *** ### Future Roadmap 🚀 **Expanded LLM support** for more AI models.\ 🚀 **Enhanced AI-driven debugging tools** for better error detection.\ 🚀 **Live collaboration features** for real-time coding with teams.\ 🚀 **More cloud deployment options** for seamless hosting.\ 🚀 **Automated testing integration** for AI-generated code validation. *** ### Conclusion AI Console is a **powerful, AI-driven development environment** that simplifies **full-stack web development** with **real-time AI assistance**. Whether you're a **developer, researcher, or AI enthusiast**, this tool provides a **seamless coding experience** with **multi-LLM support, integrated debugging, and project portability**. By following this guide, you can **effectively leverage AI Console** for **faster, smarter, and more efficient** web development. *** # Beam 2 Beta Source: https://support.myapps.ai/ultra/beam-2-beta Beam 2 Beta # Beam 2 Beta Documentation ## Table of Contents 1. [Overview](#overview) 2. [Accessing Beam 2 Beta](#accessing-beam-2-beta) 3. [Interface Layout](#interface-layout) 4. [Chat Features](#chat-features) 5. [Voice and Phone Features](#voice-and-phone-features) 6. [Persona Creator](#persona-creator) 7. [Beam Functionality](#beam-functionality) 8. [Gemini 1.5 Pro Integration](#gemini-15-pro-integration) 9. [New Features in Beam 2 Beta](#new-features-in-beam-2-beta) 10. [Settings and Preferences](#settings-and-preferences) 11. [Advanced Features](#advanced-features) 12. [Best Practices](#best-practices) 13. [Troubleshooting](#troubleshooting) 14. [FAQ](#faq) ## 1. Overview Beam 2 Beta builds upon the powerful features of its predecessor, introducing new capabilities and enhancements for an even more sophisticated AI interaction experience. \[Content for sections 2-8 remains largely the same as in the previous documentation] ## 9. New Features in Beam 2 Beta ### 9.1 File Upload Beam 2 Beta now supports a wide range of file formats for upload and analysis: ![File Upload Interface](https://img.mytsi.org/i/eR64899.png) ### 9.2 Enhanced Text Comparison Compare text with improved visualization: ![Compare Text Option](https://img.mytsi.org/i/I254886.png) ![Text Comparison View](https://img.mytsi.org/i/YquB888.png) ### 9.3 Updated Interface The preferences menu icon has been updated for better visibility: ![New Preferences Icon](https://img.mytsi.org/i/q7Ap889.png) ### 9.4 Branch Feature Create new conversation branches from existing chats: ![Branch Option](https://img.mytsi.org/i/Uraw890.png) ![Branch Creation](https://img.mytsi.org/i/M3yX892.png) ### 9.5 Split Screen Engage in multiple conversations simultaneously with split-screen functionality: ![Split Screen Feature](https://img.mytsi.org/i/Coxc895.png) ### 9.6 Prompt Caching Improved performance for the Anthropic Model through prompt caching. ### 9.7 Code Viewing and Editing View HTML and other code types directly in the chat, with options to open in external editors: ![HTML Code View](https://img.mytsi.org/i/1XNm897.png) ![Code Editor Options](https://img.mytsi.org/i/5wQS898.png) ### 9.8 Diagram Visualization Enhanced support for viewing and generating diagrams: ![Diagram View](https://img.mytsi.org/i/dD4e896.png) ### 9.9 Quick Access Shortcuts Easily accessible shortcuts above the chat prompt, including new screenshot and camera access features: ![Quick Access Shortcuts](https://img.mytsi.org/i/Aq1B887.png) ## 10. Settings and Preferences \[Content remains the same as in the previous documentation] ## 11. Advanced Features \[Previous content remains, with these additions:] * **Branch**: Create new conversation threads from existing chats * **Split Screen**: Engage in multiple conversations simultaneously * **Code Viewing and Editing**: Seamlessly view and edit code within the chat interface * **Diagram Visualization**: Enhanced support for viewing and generating diagrams ## 12. Best Practices \[Previous content remains, with these additions:] * Utilize the file upload feature for comprehensive analysis of various document types * Leverage the branch feature to explore different conversation paths * Use the split-screen functionality for multitasking and comparing different AI responses * Take advantage of the quick access shortcuts for efficient workflow ## 13. Troubleshooting \[Content remains largely the same as in the previous documentation] ## 14. FAQ \[Previous content remains, with these additions:] **Q: How does the new file upload feature work?** A: The file upload feature supports multiple file formats. Simply click the upload button, select your file, and the AI will analyze and incorporate the content into your conversation. **Q: Can I switch between branched conversations?** A: Yes, you can easily switch between branched conversations using the branch navigation feature. **Q: How does the split-screen feature enhance my workflow?** A: The split-screen feature allows you to engage in multiple conversations or tasks simultaneously, improving productivity and enabling easy comparison of different AI outputs. **Q: What types of diagrams can be visualized in Beam 2 Beta?** A: Beam 2 Beta supports various diagram types, including flowcharts, mind maps, and technical diagrams. The AI can both interpret and generate these visual representations. *** # Beam + Gemini Source: https://support.myapps.ai/ultra/beam-gemini Beam and Gemini # Beam + Gemini Documentation ## Table of Contents 1. [Overview](#overview) 2. [Accessing Beam + Gemini](#accessing-beam--gemini) 3. [Interface Layout](#interface-layout) 4. [Chat Features](#chat-features) 5. [Voice and Phone Features](#voice-and-phone-features) 6. [Persona Creator](#persona-creator) 7. [Beam Functionality](#beam-functionality) 8. [Gemini 1.5 Pro Integration](#gemini-15-pro-integration) 9. [Settings and Preferences](#settings-and-preferences) 10. [Advanced Features](#advanced-features) 11. [Best Practices](#best-practices) 12. [Troubleshooting](#troubleshooting) 13. [FAQ](#faq) ## 1. Overview Beam + Gemini is an advanced AI interaction tool that builds upon the capabilities of AI Tutor Plus. It introduces the powerful Beam functionality, allowing users to chat with multiple Language Learning Models (LLMs) simultaneously, and integrates the cutting-edge Gemini 1.5 Pro LLM with its impressive 2 million token context window. ## 2. Accessing Beam + Gemini To access the Beam + Gemini tool, users need to subscribe to the Ultra plan: ![Ultra Plan Access](https://img.mytsi.org/i/yBlv859.png) ## 3. Interface Layout The Beam + Gemini interface is designed for intuitive navigation and efficient use of its advanced features. ### 3.1 Left Sidebar * **Chat Icon**: Access chat history and create new chats ![Chat UI](https://img.mytsi.org/i/Nl2Y858.png) * Create and view chat history: ![Chat History](https://img.mytsi.org/i/HifT857.png) * Create folders for organization: ![Create Folder](https://img.mytsi.org/i/pkJY860.png) * **Voice or Phone Icon**: Access voice-related features ![Voice Features](https://img.mytsi.org/i/FEBD862.png) * **Ellipses Icon**: Access additional tools * Persona Creator * Tokenize * News about the tool ### 3.2 Top Bar (Right Side) * **Settings**: Access app settings ![Settings](https://img.mytsi.org/i/djeo843.png) * **Persona Selection**: ![Persona Selection](https://img.mytsi.org/i/raHw839.png) * **LLM Model Selection**: ![LLM Selection](https://img.mytsi.org/i/gDTc840.png) * **LLM Model Updates**: ![Update Button](https://img.mytsi.org/i/baAF841.png) ![Configure AI Models](https://img.mytsi.org/i/IuEm842.png) ## 4. Chat Features The chat interface allows for seamless interaction with AI models: * Create new chats * View chat history * Organize chats into folders ## 5. Voice and Phone Features Access voice-related features for audio interactions with AI models. ## 6. Persona Creator The Persona Creator allows users to customize AI personalities for specific tasks or interactions. ![Persona Creator](https://img.mytsi.org/i/BZQi865.png) ![Advanced Prompts](https://img.mytsi.org/i/y1SW867.png) ### 6.1 Creating Personas 1. **From YouTube Video**: ![YouTube Persona](https://img.mytsi.org/i/kXvO873.png) 2. **From Text**: ![Text Persona](https://img.mytsi.org/i/8HUD874.png) ### 6.2 Using Custom Personas 1. Generate the prompt: ![Generate Prompt](https://img.mytsi.org/i/POJd875.png) 2. Replace the system prompt: ![Replace Prompt](https://img.mytsi.org/i/9e6s876.png) ![Confirm Replacement](https://img.mytsi.org/i/fBN9877.png) 3. Test the new persona: ![Test Persona](https://img.mytsi.org/i/MYnO878.png) ## 7. Beam Functionality Beam allows users to interact with multiple LLMs simultaneously, offering diverse perspectives and insights. ### 7.1 Using Beam 1. Enter your prompt and click the Beam button: ![Beam Button](https://img.mytsi.org/i/XNuo844.png) ![Beam Activation](https://img.mytsi.org/i/kUIy845.png) 2. Select models in the Beam UI: ![Model Selection](https://img.mytsi.org/i/nVCu847.png) 3. View Beam output: ![Beam Output](https://img.mytsi.org/i/mtHJ848.png) ### 7.2 Beam Features * **Compare**: Analyze responses from different models ![Compare Feature](https://img.mytsi.org/i/2ZOA850.png) * **Guided**: Get step-by-step guidance ![Guided Feature](https://img.mytsi.org/i/kbtE853.png) ![Guided Output](https://img.mytsi.org/i/TntX854.png) * **Fuse**: Combine insights from multiple models ![Fuse Feature](https://img.mytsi.org/i/uold856.png) ## 8. Gemini 1.5 Pro Integration Beam + Gemini features integration with Google's Gemini 1.5 Pro LLM, offering: * 2 million token context window * Advanced reasoning capabilities * Improved long-form content generation ## 9. Settings and Preferences Customize your Beam + Gemini experience: * **Chat Preferences**: ![Chat Preferences](https://img.mytsi.org/i/hQGF835.png) * **Voice Preferences**: ![Voice Preferences](https://img.mytsi.org/i/uSoh836.png) * **Draw Preferences**: ![Draw Preferences](https://img.mytsi.org/i/0s3v837.png) * **Tools Preferences**: ![Tools Preferences](https://img.mytsi.org/i/PyqF838.png) ## 10. Advanced Features * **Tokenize**: Analyze token usage ![Tokenize Feature](https://img.mytsi.org/i/6N7q879.png) * **News**: Stay updated on tool improvements ![News Feature](https://img.mytsi.org/i/r9ZJ880.png) ## 11. Best Practices * Experiment with different models in Beam for diverse perspectives * Utilize custom actions to tailor outputs to your needs * Regularly check for model updates to access the latest capabilities * Use the Persona Creator to customize AI interactions for specific tasks ## 12. Troubleshooting If you encounter issues: * Ensure you're on the Ultra plan for full access * Check your internet connection * Clear browser cache and cookies * Verify that selected models are operational ## 13. FAQ **Q: How does Beam differ from regular chat interactions?** A: Beam allows you to interact with multiple AI models simultaneously, providing diverse perspectives and insights on a single query. **Q: Can I use Gemini 1.5 Pro with other models in Beam?** A: Yes, Gemini 1.5 Pro can be used alongside other models in Beam for comprehensive analysis. **Q: How do I create a custom persona?** A: Use the Persona Creator tool to generate prompts from text or YouTube videos, then apply them to your chat interactions. **Q: What's the advantage of the 2 million token context window in Gemini 1.5 Pro?** A: It allows for processing and analysis of much larger amounts of text, enabling more complex and context-rich interactions. *** # Beam Source: https://support.myapps.ai/ultra/beam-ultra Leveraging multiple AI models. ### Introduction to Beam in AI Tutor Beam in the AI Tutor app is a sophisticated feature designed to enhance your learning and research by leveraging multiple AI models. This tool allows you to input a query and receive diverse perspectives from various AI models, making it an invaluable resource for comprehensive understanding and analysis. ### Key Features of Beam * Multiple AI Models: Beam provides access to a variety of AI models, each trained with unique capabilities and knowledge bases, ensuring a wide range of responses and insights. * Interactive Tutorials: For new users, Beam includes an interactive tutorial that guides you through the process of using the feature effectively, ensuring you can leverage its capabilities fully from your first use. * Custom Actions: Beyond receiving responses, Beam allows you to combine answers, analyze, summarize, or even apply custom actions to the data, offering a highly customizable and flexible tool for your learning needs. ### Process of Using Beam in AI Tutor 1. Accessing Beam: Start by selecting the 'Ultra' tab within the AI Tutor app. 2. Inputting Prompts: Enter your question or prompt into the provided field to initiate the process. 3. Selecting Models: After submitting your prompt, you'll be directed to a page where you can choose from different AI models. Each model can provide a unique perspective or answer. 4. Reviewing Responses: Click the play button above each model to view the responses. You can engage with multiple models to gather a broad spectrum of insights. 5. Utilizing Responses: At the bottom of the page, use the toolbar to combine responses, analyze, or summarize the information. You can also add your own custom actions to further tailor the output to your needs. ### Advice for Better Practices * Explore Different Models: Each model can offer different insights based on its training and specialization. Experiment with various models to gain a more rounded understanding of your query. * Use of Custom Actions: Don't hesitate to use the custom actions to manipulate the data. This can be particularly useful for drawing connections or creating comprehensive summaries. * Regular Use of Tutorials: If you're ever unsure about how to use a feature, revisiting the tutorials can provide clarity and enhance your proficiency with the tool. ### Conclusion Beam is a powerful feature within the AI Tutor app that significantly enriches the learning experience by providing diverse insights from multiple AI models. Whether you are a student, researcher, or just curious, Beam equips you with the tools to explore complex questions with depth and breadth. By following the outlined steps and tips, you can maximize your use of this innovative feature, making your educational or research endeavors more productive and insightful. # Browser Source: https://support.myapps.ai/ultra/browser AI Tutor Browsing Agent # General Overview #### Introduction **AI Tutor Browsing Agent** is a real-time AI-powered search assistant that enables users to **fetch up-to-date information** from the web. Unlike traditional AI models that rely on pre-existing knowledge, this browsing agent can **search the internet in real time**, ensuring that responses are accurate, relevant, and current. This tool is ideal for **researchers, students, professionals, and general users** who need the latest insights, product comparisons, travel deals, and more. *** ### Getting Started with AI Tutor Browsing Agent #### **1. Accessing the Browsing Agent** To start using **AI Tutor Browsing Agent**: 1. **Open the Browsing Agent Interface** * Navigate to the **AI Tutor Browsing Agent** dashboard. * You'll see a search bar where you can type your query. 2. **Enter Your Search Query** * Type in a specific question, topic, or request. * Example queries: * *"Find top-rated Italian restaurants in NYC with outdoor seating."* * *"Compare the latest iPhone models and features."* * *"Search for London to Tokyo flight deals."* * *"Explore recent renewable energy innovations."* 3. **Submit Your Query** * Click the **send button (paper plane icon)** to initiate the search. * The AI will browse the internet, retrieve relevant information, and present a structured response. *** ### How It Works AI Tutor Browsing Agent follows a **multi-step process** to fetch and present information: #### **1. Query Processing & Optimization** * The AI **understands the intent** behind your search. * It reformulates the query if necessary to retrieve the most relevant results. #### **2. Real-Time Web Search** * The AI performs a **live internet search** using trusted sources. * It pulls the most recent and **highly relevant** data. #### **3. Data Extraction & Summarization** * The AI **filters out irrelevant or outdated information**. * It **summarizes key points** for easy understanding. #### **4. Structured Response Generation** * The AI presents the findings in a **well-organized format**. * It may include: * **Bullet points** for quick insights. * **Comparisons** for product research. * **Links to sources** for further reading. *** ### Key Features & Functionalities #### **1. Real-Time Internet Search** * Retrieves the **latest** information directly from the web. * Ensures **up-to-date** responses for time-sensitive queries. #### **2. Smart Query Optimization** * Enhances search accuracy by **restructuring vague queries**. * Finds the **best sources** for reliable information. #### **3. Structured & Summarized Responses** * Provides **concise, well-organized answers**. * Highlights **key points** and **important takeaways**. #### **4. Predefined Search Suggestions** * Offers **quick-access search prompts** for common queries. * Includes topics like: * **Restaurant recommendations** * **Product comparisons** * **Travel deals** * **Technology trends** #### **5. Source Citations (If Available)** * Some results may include **links to original sources**. * Allows users to verify and explore further. *** ### Best Practices for Effective Browsing To maximize the effectiveness of **AI Tutor Browsing Agent**, follow these tips: ✅ **Be specific** with your search queries for more relevant results.\ ✅ **Use comparative queries** (e.g., *"Compare iPhone 14 vs. iPhone 15"*) for better insights.\ ✅ **Check source links** when available to verify information.\ ✅ **Use follow-up searches** to refine results further.\ ✅ **Avoid overly broad queries**—narrowing the focus improves accuracy. *** ### Future Enhancements & Roadmap 🚀 **Advanced filtering options** to refine search results.\ 🚀 **Personalized search preferences** for customized browsing.\ 🚀 **Integration with academic sources** for scholarly research.\ 🚀 **AI-powered news aggregation** for trending topics.\ 🚀 **Multi-source comparisons** for products, travel, and more. *** ### Conclusion AI Tutor Browsing Agent is a **powerful, real-time search assistant** that helps users **fetch, analyze, and summarize** the latest information from the web. Whether you're looking for **product comparisons, travel deals, research insights, or general knowledge**, this tool delivers **accurate, structured, and up-to-date responses**. By following this guide, you can **effectively use AI Tutor Browsing Agent** to enhance your research, decision-making, and information gathering. *** # Deep Research Source: https://support.myapps.ai/ultra/deep-research AI Tutor Deep Research # General Overview #### Introduction **AI Tutor Deep Research** is an advanced AI-powered research assistant that allows users to generate **comprehensive reports** on any topic. By leveraging cutting-edge AI models, it performs deep, structured research to provide **well-organized, insightful, and data-driven** results. This feature is designed for **students, researchers, and professionals** who need in-depth analysis, summaries, and structured insights without manually sifting through vast amounts of information. *** ### Getting Started with AI Tutor Deep Research #### **1. Accessing Deep Research** To begin using **AI Tutor Deep Research**: 1. **Navigate to the Deep Research Dashboard** * Open the **AI Tutor Deep Research** interface. * You'll see a simple prompt asking, *"What would you like to research?"* 2. **Enter Your Research Query** * Type in any topic, question, or specific area of interest. * The AI will analyze and structure the research based on your input. 3. **Adjust Research Parameters** * **Breadth:** Determines how wide the research scope is. * **Depth:** Controls how detailed and in-depth the research will be. * **Model Selection:** Choose from different AI models for varying levels of complexity and accuracy. 4. **Start the Research Process** * Click the **submit button** to begin generating a structured research report. *** ### Research Methodology & AI Capabilities AI Tutor Deep Research utilizes a **multi-step research methodology** that ensures relevant, well-structured, and accurate information: #### **1. Contextual Understanding & Query Expansion** * The AI **interprets** your input and expands the query to gather more relevant information. * It **identifies key themes, subtopics, and related concepts** to structure the research effectively. #### **2. Information Retrieval & Data Processing** * The AI pulls **verified** and **contextually relevant** data from trusted sources. * It applies **semantic search techniques** to retrieve the most accurate information. #### **3. Structured Report Generation** * The AI organizes the findings into a **coherent, well-structured report**. * It includes: * **Summaries** of key findings. * **Bullet points & categorized sections** for easy readability. * **Citations & references** (if applicable). #### **4. Depth & Breadth Control** * Users can **customize** the depth and breadth of research to match their needs. * **Shallow research** provides quick overviews, while **deep research** offers extensive analysis. *** ### Key Features & Functionalities #### **1. Adjustable Research Scope** * **Breadth Control:** Expands or narrows the research coverage. * **Depth Control:** Adjusts how detailed the AI’s response will be. #### **2. AI-Powered Summary & Insights** * Generates **concise summaries** for quick understanding. * Highlights **key takeaways** and important findings. #### **3. Multi-Model Support** * Users can select different AI models for research, balancing **speed, accuracy, and detail**. #### **4. Interactive Research Experience** * Users can **refine queries** and **ask follow-up questions** to improve the quality of research. #### **5. Comprehensive Report Export** * Users can **download** or **copy** the generated research for future reference. * Export options include **text, markdown, and PDF formats**. *** ### Best Practices for Effective Research To get the most out of **AI Tutor Deep Research**, follow these best practices: ✅ **Be specific** with your research queries for more targeted results.\ ✅ **Adjust depth and breadth** based on the level of detail needed.\ ✅ **Use follow-up questions** to refine the research further.\ ✅ **Review and verify** AI-generated insights, especially for academic or professional use.\ ✅ **Utilize different models** to compare research outputs. *** ### Future Enhancements & Roadmap 🚀 **Multi-source research aggregation** for more diverse perspectives.\ 🚀 **Real-time citation generation** for academic research.\ 🚀 **Customizable report templates** for different use cases.\ 🚀 **Integration with external databases** for more authoritative sources.\ 🚀 **AI-driven summarization of long documents and PDFs.** *** ### Conclusion AI Tutor Deep Research is a **powerful AI-driven tool** that simplifies the research process by delivering **structured, insightful, and in-depth reports** on any topic. Whether you're a **student, researcher, or professional**, this tool helps you **save time, improve accuracy, and enhance your understanding** of complex subjects. By following this guide, you can **maximize the potential** of AI Tutor Deep Research and **streamline your research workflow** effectively. *** \#AITutor #DeepResearch #AIReports #NoCodeResearch #AIInsights # Glyph Source: https://support.myapps.ai/ultra/glyph Advanced AI-powered voice and vision assistant. ![AI Tutor GLYPH](https://img.mytsi.org/i/ImvN966.png) # General Overview #### Introduction **AI Tutor GLYPH** is an advanced **voice and vision AI assistant** that enables users to **converse naturally, analyze images, and interact with AI in a more immersive way**. With cutting-edge **speech recognition, text-to-speech synthesis, and multimodal AI capabilities**, GLYPH enhances how users interact with AI in their daily lives. GLYPH allows users to **activate AI with a play button**, **listen to AI-generated responses**, **download audio files**, and **view transcriptions** for better accessibility and usability. This guide provides an **overview of GLYPH’s features, setup instructions, and best practices** for voice and image-based interactions. *** ### Key Features #### **1. Activate Glyph with the Play Button** * **Press the play button** to **start and activate Glyph**. * Once activated, Glyph **begins listening and responding in real-time**. * Enables **hands-free AI interaction** with voice input. #### **2. Voice Conversations with AI** * Engage in **real-time back-and-forth conversations** with AI. * Request **stories, explanations, or assistance** through voice commands. * AI responds with **natural, human-like speech**. #### **3. Image Recognition & Analysis** * Upload or capture images for **real-time AI analysis**. * Use AI to **identify objects, troubleshoot issues, or analyze data**. * Draw on images to **highlight specific areas for AI focus**. #### **4. Advanced Text-to-Speech (TTS)** * AI generates **realistic voice responses** from text. * Choose from **multiple AI-generated voices** for a personalized experience. * Built using **professional voice actors and AI synthesis models**. #### **5. Speech Recognition & Transcription** * Convert **spoken words into text** using advanced **speech-to-text models**. * Supports **natural language understanding** for fluid conversations. * Enables **hands-free AI interaction**. #### **6. Play Button, Audio Download & Transcription** * **Press the play button** to **start Glyph** and engage in voice AI. * **Download audio files** for offline listening or sharing. * **View transcriptions** of AI responses for accessibility and reference. #### **7. Multimodal AI for Voice & Vision** * Combines **GPT-4’s language capabilities** with **image understanding**. * AI can **interpret photos, documents, and screenshots**. * Provides **context-aware responses** based on images and text. #### **8. Mobile & Cross-Platform Support** * Available on **iOS, Android, and web platforms**. * Easily **switch between voice and text interactions**. * Optimized for **touchscreen and voice-controlled devices**. *** ### Getting Started with AI Tutor GLYPH #### **1. Activating Glyph with the Play Button** * **Press the play button** to **start and turn on Glyph**. * Once Glyph is active, it **listens for voice input** and responds in real-time. * You can **pause or stop Glyph** at any time. #### **2. Enabling Voice Conversations** * Open the **AI Tutor GLYPH** app on **iOS or Android**. * Navigate to **Settings → New Features**. * Enable **Voice Conversations**. * Tap the **headphone icon** on the home screen to start a voice chat. #### **3. Choosing a Voice** * Select from **five available AI voices**. * Voices are generated using **text-to-speech AI models**. * Each voice is designed to sound **natural and expressive**. #### **4. Using Image-Based AI Features** * Tap the **photo button** to capture or upload an image. * On mobile, tap the **plus button** first to access image options. * Draw on the image to **highlight specific areas for AI analysis**. #### **5. Playing, Downloading, and Transcribing Audio** * **Press the play button** to **activate Glyph and start voice AI**. * **Download audio files** for offline playback. * **View transcriptions** for easy reading and reference. #### **6. Real-Time AI Assistance** * Ask AI to **describe, analyze, or troubleshoot** based on images. * Use AI to **generate captions, extract text, or provide insights**. * AI can process **photographs, screenshots, and complex visual data**. *** ### How AI Tutor GLYPH Works #### **1. Voice Processing & Text-to-Speech** * AI converts **spoken words into text** using **Whisper AI**. * AI generates **human-like speech responses** using **text-to-speech synthesis**. * Built with **deep learning models trained on real human voices**. #### **2. Image Understanding & AI Analysis** * AI applies **GPT-4’s reasoning abilities** to process images. * Recognizes **objects, text, graphs, and complex visual elements**. * AI can **answer questions based on images** and **provide step-by-step guidance**. #### **3. Multimodal AI Capabilities** * Combines **speech, text, and vision** into a unified AI experience. * AI can **respond to both voice and image-based inputs**. * Enables **interactive, real-world AI assistance**. *** ### Best Practices for Using AI Tutor GLYPH To maximize your experience with **AI Tutor GLYPH**, follow these best practices: ✅ **Press the play button** to **activate Glyph** before speaking.\ ✅ **Use clear voice commands** for better AI understanding.\ ✅ **Choose the right AI voice** for a more engaging experience.\ ✅ **Highlight key areas in images** using the drawing tool.\ ✅ **Ask follow-up questions** to refine AI responses.\ ✅ **Use AI for troubleshooting, research, or creative assistance.**\ ✅ **Download audio files** for easy access to AI-generated responses.\ ✅ **Check transcriptions** for better readability and reference. *** ### Future Roadmap 🚀 **Expanded voice options** with customizable AI voices.\ 🚀 **Live translation capabilities** for multilingual conversations.\ 🚀 **Improved AI-driven image editing & annotation tools.**\ 🚀 **Integration with smart home assistants & IoT devices.**\ 🚀 **Enhanced real-time speech synthesis for more natural conversations.** *** ### Conclusion AI Tutor GLYPH is a **powerful AI assistant** that combines **voice, vision, and language understanding** into a seamless, interactive experience. Whether you're **having a conversation, analyzing images, or troubleshooting issues**, GLYPH provides **intelligent, real-time AI assistance**. With **playback, audio downloads, and transcriptions**, GLYPH offers **greater flexibility and accessibility** for users who want to engage with AI in new ways. By following this guide, you can **make the most of AI Tutor GLYPH** and experience **next-generation AI interactions**. *** # Merging Responses in Beam Source: https://support.myapps.ai/ultra/merging-in-beam Merging Responses in Beam # Merging Responses In Beam ### Understanding the Merge Toolbar in Your Application The merge toolbar at the bottom of your interface is designed to enhance how you consolidate and refine AI-generated responses. This tool offers several options---Fuse, Guided, Compare, and Custom---each tailored to different needs and preferences. Here's a detailed look at how to utilize these features effectively: #### Fuse * Usage: Ideal for integrating conversation details and ideas into a unified, clear response. * Process: Click on "Add Fusion" to start merging content. Note that while Fuse is excellent for textual content, it may not be as effective for coding-related responses. #### Guided * Usage: Perfect for selectively curating information from various responses. * Process: Select "Guided" and then click "Add a Checklist". This enables you to create a checklist from your responses, allowing you to select specific information to retain. Confirm your selections to generate a new, consolidated response based on your choices. #### Compare * Usage: Useful for a detailed analysis and comparison of different AI responses. * Process: Choose "Compare" and then "Add Breakdown". This function analyzes and compares responses, presenting them in a structured, empirical format that aids in making informed decisions about which elements to include in your final response. #### Custom * Usage: Best for highly specific and tailored response merging. * Process: Use the Custom option to input your own merging prompt, which will guide the tool in combining all answers into a cohesive response according to your precise requirements. #### Additional Tools and Settings * Merge Model Selection: Next to the merge toolbar, a dropdown allows you to select which model you'd like to use for merging your responses. Feel free to experiment with different models to find the one that best suits your needs. * Managing Responses: After merging, your original responses will remain accessible. You can remove any response using the minus button located at the top right corner of each response window. Additionally, if you change the merging model, you might want to redo the answer to reflect the new model's methodology. ### Conclusion The merge toolbar is a powerful feature that brings flexibility and precision to how you handle AI-generated content. By understanding and utilizing the Fuse, Guided, Compare, and Custom options, you can tailor your responses to be as informative and effective as possible. Whether you're looking to create a comprehensive answer, select the best information, or compare insights, these tools provide the necessary functionality to enhance your decision-making and output quality. # Multibot Ultra - Adding Assistants Source: https://support.myapps.ai/ultra/multibot-ultra-adding-assistants Multibot Ultra - Adding Assistants ### Introduction Multibot Ultra, accessible via AI Tutor, offers a robust platform for creating and managing a diverse range of virtual assistants tailored to specific operational needs. This comprehensive guide will walk you through the process of adding and customizing assistants within Multibot Ultra, enhancing functionality with various integrative tools and features. #### Key Points 1. Accessing Multibot Ultra: * Login: Start by logging into the Ultra section of AI Tutor using your credentials or by continuing with Google. * Dashboard: Upon login, you will be directed to the Multibot Ultra dashboard, which serves as your central hub for managing and creating assistants. 2. Creating and Customizing Assistants: * Navigate to Assistants: Use the dropdown menu at the top left corner to select 'Assistants✨' , which opens the assistant management sidebar. ![](https://docs.myapps.ai/storage/page_media/cf92938d-48e2-4553-af84-b61c3e30da94.png) * Add a New Assistant: Click to add a new assistant, starting the customization process. * Personalization: * Name and Picture: Assign a unique name and upload a representative image for your assistant. * Description and Instructions: Provide a detailed description and specific instructions delineating the assistant's functions and responsibilities. * AI Model and Features: * Model Selection: Choose the appropriate AI model that aligns with the tasks your assistant will perform. * Code Interpreter: Consider enabling the code interpreter to allow your assistant to handle and execute code-based queries. 3. Integrating Tools and Services: * Tools Configuration: Access the 'Tools' section to explore and configure available integrations. * Service Selection: Here you can choose from various services to integrate into your assistant. Examples include Google API and DALL-E 3, which enable functionalities like web searching and image generation, respectively. ![](https://docs.myapps.ai/storage/page_media/9e6d3114-13d2-4caf-983c-a20b00b8d028.png) * DALL-E 3 Example: When enabled, DALL-E 3 allows your assistant to generate images based on textual descriptions, expanding its creative capabilities. ![](https://docs.myapps.ai/storage/page_media/a040e67d-16e3-4bf2-b0a4-f0703aa9d78a.png) * Customize Based on Needs: Add services based on the specific actions you need your assistant to perform. Some of these services may require API keys to function properly. 4. File Management: * Access Files: Manage files associated with each assistant through the 'Files' tab. * Organize and Utilize: Download, update, or remove files as necessary to keep your assistant's resources up-to-date and relevant. 5. Managing Multiple Assistants: * Seamless Switching: Effortlessly create and switch between multiple assistants, each designed for distinct tasks or environments, using the same user interface. * Efficient Management: Adjust settings, update capabilities, or deactivate assistants as your operational needs evolve. #### Conclusion Multibot Ultra empowers users to fully leverage the capabilities of virtual assistants, making it easy to create, customize, and manage multiple assistants with diverse functionalities. By integrating a variety of tools and services, you can tailor each assistant to meet specific needs, from routine tasks to complex problem-solving. This flexibility, combined with an intuitive setup process, ensures that you can maximize productivity and enhance interactions across various platforms and contexts. Explore the possibilities with Multibot Ultra and transform the way you engage with digital environments. # Video Editor Source: https://support.myapps.ai/video-editor A real timeline for assembling, captioning, and rendering video The **Video Editor** is a timeline. Bring in clips, arrange them, add captions, and render — the finishing step after generation. ## When to use it | Surface | Role | | ------------------------------------------ | ----------------------------------------------------------- | | **[Generate](/generate)** | Make one clip | | **[Video Agent](/create/video-agent)** | Generate a multi-segment episode with character consistency | | **[Cinema Studio](/create/cinema-studio)** | Plan and direct shots before generating | | **Video Editor** | Assemble, trim, caption, and render the final cut | Reach for the editor when you have footage and need precise control over timing. ## Projects Work is organized into projects, each with a name, optional description, and thumbnail. * **Create** a project and name it * **Search** your projects by name * **Open** one to edit * Changes save as you work ## Editing Inside a project you get a **timeline** where clips are arranged and timing controlled, plus tools to add clips, edit captions, and trim. Source material comes from your generations and uploads in [My Assets](/assets) — generate in Pixio, then assemble here. ## Rendering Rendering produces the final video file from your timeline. Rendering happens **in the app**. Headless renders, exports, and render progress aren't available through the API — you can build and edit a timeline programmatically, but the final render is app-only. See [what you can automate](/integrations/what-you-can-automate). ## Working efficiently Fixing a bad clip is a new generation, not an edit. Get the clips right before building the timeline around them. Everything on the timeline comes from your library — it's the staging area. Captions follow timing. Re-cutting after captioning means redoing them. Rendering is the commit step. Preview before you spend the time. Timeline project CRUD and validated editor operations are scriptable — only the render isn't. Useful if you want to construct a timeline from code and finish it in the app. # Pixio Maker Mode Source: https://support.myapps.ai/video/maker-mode Unlimited Generations # Unlimited Generations with MotionCraft Ultra and VidCraft Ultra If your account has our early bird pricing, canceling your subscription will eliminate this discount and you will be paying the normal rate when you resubscribe. Unleash your creativity with unlimited AI-powered video and motion graphics generation! ## Overview Maker Mode is a revolutionary feature that allows users to harness the full potential of MotionCraft Ultra and VidCraft Ultra with unlimited generations. This powerful mode enables creators to experiment, iterate, and perfect their AI-generated videos and motion graphics without constraints. Create stunning motion graphics and animations with AI-powered tools. Generate high-quality videos from text, images, and existing videos using advanced AI. ## Key Features of Maker Mode * **Unlimited Generations**: Create as many iterations as you need without restrictions. * **Rapid Prototyping**: Quickly test and refine your ideas with instant feedback. * **Enhanced Creativity**: Explore diverse concepts and styles without limitations. * **Improved Workflow**: Streamline your creative process with unrestricted access to AI tools. ## How to Access Maker Mode Maker Mode requires a Maker Mode subscription. You can now upgrade to Maker Mode and it is \$159.99 Monthly. Once payment is complete, Maker Mode will be activated in your account. To upgrade or for any questions about Maker Mode, please contact [rcohen@mytsi.org](mailto:rcohen@mytsi.org). ## Unleashing Creativity with MotionCraft Ultra MotionCraft Ultra in Maker Mode allows you to push the boundaries of motion graphics creation. ### Key Advantages 1. **Endless Iterations**: Refine your animations until they're perfect. 2. **Style Exploration**: Experiment with various visual styles and techniques. 3. **Complex Sequences**: Build intricate motion graphics by combining multiple generations. Use Maker Mode to generate multiple layers of motion graphics, then combine them for rich, complex compositions. ## Maximizing VidCraft Ultra Potential With Maker Mode, VidCraft Ultra becomes an even more powerful tool for video creation. ### Enhanced Capabilities 1. **Extended Video Durations**: Create longer, more complex video sequences. 2. **Style Refinement**: Iterate on video styles until you achieve the perfect look. 3. **Content Variations**: Generate multiple versions of a video to choose the best one. Use Maker Mode to create multiple versions of your video with slight variations, then test them with your audience to determine the most effective one. ## Conclusion Maker Mode transforms MotionCraft Ultra and VidCraft Ultra into limitless creative tools. By removing generation restrictions, it empowers creators to explore, experiment, and produce their best work without constraints. Upgrade to Maker Mode today and take your AI-powered video and motion graphics creation to the next level! # Workflows Source: https://support.myapps.ai/workflows Chain AI models into multi-step flows. Build once, run with one click. **Workflows** let you chain **multiple AI models** into a single flow. You build the flow once (add models, connect them, set prompts and options), then **Run** to execute every step in order. Outputs from one step automatically feed into the next. No need to run each model by hand or copy URLs between steps. Workflows are **BETA**. The builder and behavior may evolve. *** ## Where to find Workflows * In the app sidebar, click **Workflows** (BETA). * You’ll see a list of your workflows and a **Create workflow** button. *** ## Workflow list * **Cards** show: name, description, last run time, status (queued / running / succeeded / failed), and a preview of the final output (image, video, or “3D model” placeholder). * **Actions** on each card: * **Edit** — Open the workflow in the builder. * **Download** — Download the latest output file (if any). * **Publish** — Share the workflow via a public link (optional cover image). * **Delete** — Delete the workflow and all its runs. * **Realtime**: When a run is queued or running, the list updates automatically; the card shows a “running” state and you can open the workflow to watch step-by-step progress. *** ## Creating a workflow 1. Click **Create workflow**. 2. Enter a **name** and optional **description**, then confirm. You’re taken to the **workflow builder** with an empty canvas. 3. In the builder: * **Left panel — Models**: Search and browse by provider/company. Use **+** on a model to add it to the canvas. You can star models as favorites for quick access. * **Right panel — Save & Run**: Set or change name/description, **Save** the workflow, and **Run** it. After a run, status and “Started/Finished” time are shown here; the canvas also updates in real time. *** ## How the builder works ### Adding steps * **Only “model” nodes** are added from the library. Each node is one AI model (e.g. text-to-image, image-to-video, upscale, TTS). * Click **+** next to a model in the left panel to add it to the canvas. You can add as many nodes as you need. * Each model node shows: * **Label** (model name) * **Inputs** — e.g. prompt, image URL, video URL, duration, model-specific options. Inputs can be: * Filled **manually** (type prompt, pick a file from your assets, set sliders/selects). * **Connected** from another node’s output (drag from the **result** handle of one node to a **file** handle on another). ### Connecting nodes * **Valid connection**: From a node’s **output (result)** handle → to another node’s **input (file)** handle. * **Type compatibility** is enforced: * **Image** output can only connect to **image** inputs. * **Video** → **video**, **audio** → **audio**. * If you try to connect incompatible types (e.g. image to video), you’ll see an error and the connection is rejected. * **Multi-input**: Some models accept multiple files (e.g. several images). You can connect multiple nodes to the same input up to that input’s maximum. * **Execution order** is determined automatically from the graph: nodes with no incoming connections run first; then nodes that depend on their outputs, and so on. Cycles are not allowed and will cause an error when you run. ### What gets saved * **Saved with the workflow**: Node positions, which model each node uses, **prompt** and **negative prompt**, **params** (sliders, selects, etc.), **file URLs** you picked or that came from previous runs, and **connections** (edges). So when you run again, the same inputs and options are used unless you change them. * **Not persisted as “workflow definition”**: Transient run state (e.g. “running”) is cleared after a run; the builder may auto-save after a run to store the latest output URLs on nodes for preview. ### Locked nodes * You can **lock** a model node when it already has an output you’re happy with. When the workflow runs, **locked** nodes are **skipped** (no new API call); their existing output is reused for any downstream node connected to them. Use this to avoid re-running expensive steps while you iterate on later steps. *** ## Running a workflow 1. **Save** first (especially if you just created the workflow or changed name/description). **Run** will auto-save the current definition if needed, then start a new run. 2. Click **Run**. A new **workflow run** is created (status: queued → running). 3. **Execution**: * Steps run in **topological order** (no step runs until the steps that feed it have finished). * Each step uses: the model’s params, prompt/negative prompt from the node, and any **file/image/video/audio URLs** from connected nodes (or from the node’s own selected assets). * **Credits** are consumed per step according to each model’s pricing; duration-based models (e.g. video, audio) get **duration** inferred from upstream outputs when possible. 4. **Realtime updates**: The builder subscribes to the current workflow’s runs. As each step completes, the corresponding node updates (e.g. “Generation completed”, output thumbnail/URL). The **Save & Run** panel shows overall status and “Started” / “Finished” time. 5. When the run **succeeds** or **fails**, the workflow is marked complete. If it succeeds, the last step’s output (or the rightmost node with output, depending on layout) is what you see as the “final” result in the list and in the builder. You can **download** that output from the list or from the node in the builder. *** ## After a run * **List**: The workflow card shows the latest run status and a preview of the final output (image/video/3D). You can **Download** to get the file. * **Builder**: Each model node that ran shows its output (image/video/audio, etc.). You can lock nodes, change prompts/params, add or remove nodes, and **Run** again. * **Publish**: From the list or the builder, you can **Publish** the workflow to get a shareable link. Others can view (and optionally duplicate) the workflow; they don’t edit your copy unless they duplicate it to their account. *** ## Tips and constraints * **Start simple**: One text-to-image model, then add an upscale or image-to-video model and connect the first node’s output to the second. Run to see the full chain. * **Prompt and files**: The “first” step(s) in the graph usually need prompt or file inputs set manually (or from your assets). Downstream steps can rely entirely on connected outputs. * **Duration**: For video/audio/lipsync models, the runner tries to pass **duration** from upstream outputs so credit calculation and APIs that need length work correctly. * **Errors**: If a step fails, the run stops and is marked failed; you’ll see which node failed and the error. Fix inputs or model choice and run again. * **No cycles**: The graph must be acyclic. If you create a cycle by mistake, running will fail with a “Cycle detected” error. * **Models available**: Almost all app models appear in the workflow model library (by provider/company), except a few (e.g. some LoRA/training, certain extend-only or internal models). Search by name, type, or company to find what you need. *** ## Summary | What | How | | ------------- | ------------------------------------------------------------------------------------------------ | | **Open** | Sidebar → Workflows | | **Create** | Create workflow → name/description → builder | | **Add steps** | Models panel → + on a model → node on canvas | | **Connect** | Drag from a node’s **result** handle to another’s **file** handle (same type: image/video/audio) | | **Configure** | Set prompt, params, and/or file inputs on each node; lock nodes to reuse existing output | | **Save** | Save & Run panel → Save (or auto-save after run) | | **Run** | Save & Run panel → Run → watch status and node updates in real time | | **Result** | Final output on last/downstream node; download from list or use in next run | | **Share** | Publish → get link; others can view or duplicate | *** ## How it works (deep dive) This section maps the product behavior to how the system actually runs workflows. Useful if you’re debugging or integrating. ### Data model * **Workflow**: Stored in `workflows` with `id`, `name`, `description`, `definition` (JSON). `definition` has `version: 1`, `nodes[]`, `edges[]`. * **Node**: `id`, `type` (`input` | `file` | `model` | `output`), `position`, `data` (prompt, fileUrl, providerId, modelId, params, outputUrl, locked, etc.). In the builder you only add **model** nodes; input/file/output types exist for the runner and for read-only views. * **Edge**: `source`, `target`, optional `sourceHandle` (e.g. `result`, `result-0`), `targetHandle` (e.g. `file-0`). Connections are from **result** → **file-N** (N = input index on the target model). ### Execution 1. **Start run**: User clicks Run → `startWorkflowRunAction` creates a row in `workflow_runs` (status `queued`) with a **snapshot** of the current `definition`, then invokes the **workflow orchestrator** (e.g. Edge Function or internal API) with the run ID. 2. **Orchestrator** calls `runWorkflow(runId)`: loads the run and its `definition_snapshot`, computes **execution order** via a topological sort of nodes by edges (cycles throw). 3. **Per node** (in order): * **input** node: contributes `prompt` / `negativePrompt` to params of nodes connected to it. * **file** node: contributes `fileUrl` (and optional `durationSeconds`) to connected nodes. * **model** node: If **locked** and has `outputUrl`, skip call and push existing output into the run’s `outputs` map. Otherwise **prepare** params with `buildParamsFromGraph` (merge node’s `params` + prompt/negative + all inbound edge sources: input prompt, file URLs, or upstream model `outputUrl`/duration), validate with the model’s schema, resolve credits, then **dispatch** generation (the same models as the main app). On success, store `outputUrl` (and optional `outputs`/duration) in the run’s `outputs` and in the step record. * **output** node: no execution; just marks where the “final” output comes from (the connected source). 4. **Progress**: After each step, the run’s `steps` array and `updated_at` are written to the DB. The app subscribes to **realtime** (`workflow_runs` filtered by `workflow_id`) so the UI updates as steps complete. 5. **End**: Run status set to `succeeded` or `failed`; optional `error_message`. On success, the workflow’s `definition` can be updated so the last node’s `outputUrl` is persisted for preview in the list. ### Connection rules (runner) * **buildParamsFromGraph** resolves inbound edges by `targetHandle` (e.g. `file-0` → first file input). Source kinds: `input` (prompt/negativePrompt), `file` (fileUrl + duration if media), `model` (outputUrl + duration if audio/video). Params are merged; multi-file inputs get arrays when the model supports it. * **Type compatibility** in the builder: output type (image/video/audio) is inferred from `modelType` (e.g. “text-to-video” → video); input type from the model’s input definition. Connection allowed only when types match. ### Credits and duration * Each model step is validated and **credits** are computed from the model’s pricing (and plan). Duration-based pricing uses **duration** from: upstream node’s `outputs`/params, or probing the media URL when the step is file/audio/video. * Total cost of a run is the sum of each executed (non-locked) step’s credits.