Skip to main content

​
Overview

This endpoint provides a streaming chat mechanism that works seamlessly with the Vercel AI SDK. After creating a token for a user, you can leverage this endpoint to power real-time chat interfaces using frameworks like React and the Vercel AI SDK’s useChat hook.

​
Endpoint

  • HTTP Method: GET
  • URL: /chat/{token}/stream
Replace {token} with the actual chat token assigned to the user.

​
Features

  • Real-time streaming: Stream chat messages as they are generated.
  • Vercel AI SDK Compatible: Integrates easily with tools like ai/react for a reactive chat interface.
  • Robust Error Handling: Option for keeping the last message on error.

​
Request

Path Parameter
NameRequiredTypeDescription
tokenYesstringThe unique chat session token assigned to the user
HTTP Headers
  • accept: application/json
No body content is required for this GET request.

​
Example Request

curl --request GET \
     --url "https://aitutor-api.vercel.app/api/v1/chat/your_chat_token/stream" \
     --header 'accept: application/json'

​
Response

The response will be a real-time stream of chat messages. The format is dynamically generated with each message containing the following fields: role: Identifies the sender (e.g., user, assistant). content: The text/content of the message. timestamp: ISO 8601 formatted timestamp marking when the message was sent. Code Example Below is an example of how to use this endpoint in a React component with the Vercel AI SDK’s useChat hook: 

'use client';

import { useChat } from 'ai/react';

export default function Page() {
  // Replace 'your_chat_token' with the actual token value for the chat session
  const token = 'your_chat_token';
  
  // Use the hook and specify the API endpoint for streaming chat
  const { messages, input, handleInputChange, handleSubmit } = useChat({
    api: `https://aitutor-api.vercel.app/api/v1/chat/${token}/stream`,
    keepLastMessageOnError: true,
  });

  return (
    <>
      {messages.map(message => (
        <div key={message.id}>
          {message.role === 'user' ? 'User: ' : 'AI: '}
          {message.content}
        </div>
      ))}

      <form onSubmit={handleSubmit}>
        <input name="prompt" value={input} onChange={handleInputChange} />
        <button type="submit">Submit</button>
      </form>
    </>
  );
}

​
Additional Notes

Vercel AI SDK Integration: For more details on integration with Vercel AI SDKs, visit the Vercel AI SDK documentation. Error Handling: The keepLastMessageOnError option ensures that if an error occurs while streaming, the last successful message is preserved until the issue is resolved. Customization: Feel free to customize the chat UI and error handling logic according to your specific requirements.
⌘I