Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/Wikedhart18/nextjs-ai-chatbot/llms.txt

Use this file to discover all available pages before exploring further.

All endpoints require an active session cookie obtained via the authentication flow. Unauthenticated requests receive a 401 Unauthorized response.

​
POST /api/chat

Send a message and stream the AI response back to the client. If the chat ID does not yet exist in the database, a new chat record is created automatically with a generated title. The response is a AI SDK data stream (text/plain transfer-encoding chunked). The stream includes text deltas, reasoning tokens (when the reasoning model is selected), and tool-call results. Authentication: required (session cookie)

​
Request body

​
id
string
required
UUID of the chat conversation. Generate a new UUID on the client when starting a fresh conversation.
​
messages
Message[]
required
Full conversation history in AI SDK Message format. The last entry must be a user message.
​
selectedChatModel
string
required
The model identifier to use for this turn. One of:
ValueDescription
chat-model-smallFast, lower-cost model
chat-model-largeHigh-capability model
chat-model-reasoningExtended reasoning model (tools disabled)

​
Response

A streaming text/plain data stream in the AI SDK wire format. The stream carries:
  • Text deltas — incremental assistant text tokens.
  • Reasoning tokens — present when selectedChatModel is chat-model-reasoning.
  • Tool results — for getWeather, createDocument, updateDocument, and requestSuggestions (disabled for the reasoning model).

​
Error codes

StatusMeaning
400No user message found in the provided messages array.
401Missing or invalid session.
500Internal error during streaming (the stream body will contain "Oops, an error occured!").

​
Example

curl
curl -X POST https://your-app.vercel.app/api/chat \
  -H 'Content-Type: application/json' \
  -H 'Cookie: authjs.session-token=<token>' \
  --no-buffer \
  -d '{
    "id": "550e8400-e29b-41d4-a716-446655440000",
    "messages": [
      { "id": "msg_1", "role": "user", "content": "What is the weather in Paris?" }
    ],
    "selectedChatModel": "chat-model-small"
  }'
TypeScript (AI SDK)
import { useChat } from 'ai/react';

const { messages, input, handleInputChange, handleSubmit } = useChat({
  api: '/api/chat',
  body: {
    selectedChatModel: 'chat-model-small',
  },
  id: chatId,
});

​
DELETE /api/chat

Permanently delete a chat and all of its messages and votes. Authentication: required (session cookie). You can only delete chats that belong to your user account.

​
Query parameters

​
id
string
required
UUID of the chat to delete.

​
Response

Returns 200 Chat deleted as plain text on success.

​
Error codes

StatusMeaning
401Missing or invalid session, or the chat belongs to a different user.
404No id query parameter was provided.
500Unexpected database error.

​
Example

curl
curl -X DELETE \
  'https://your-app.vercel.app/api/chat?id=550e8400-e29b-41d4-a716-446655440000' \
  -H 'Cookie: authjs.session-token=<token>'
TypeScript
await fetch(`/api/chat?id=${chatId}`, { method: 'DELETE' });

​
POST /api/files/upload

Upload an image file to Vercel Blob storage. The returned URL can be passed as an image content part inside a messages array to give the model visual context. Authentication: required (session cookie)

​
Request body

Multipart form-data with the following field:
​
file
File
required
The image to upload. Constraints:
  • Maximum size: 5 MB
  • Accepted types: image/jpeg, image/png

​
Response

Returns the Vercel Blob PutBlobResult object as JSON.

​
Error codes

StatusMeaning
400No file provided, empty request body, file exceeds 5 MB, or unsupported file type.
401Missing or invalid session.
500Vercel Blob upload failed or request could not be processed.

​
Example

curl
curl -X POST https://your-app.vercel.app/api/files/upload \
  -H 'Cookie: authjs.session-token=<token>' \
  -F 'file=@/path/to/image.png'
TypeScript
const formData = new FormData();
formData.append('file', file);

const response = await fetch('/api/files/upload', {
  method: 'POST',
  body: formData,
});

const { url } = await response.json();

Build docs developers (and LLMs) love

Get started for freeTalk to us