File: render-visual-interface-in-chat.md | Updated: 11/15/2025
Menu
Google Gemini Image Generation
Get started with Claude 3.7 Sonnet
Get started with OpenAI o3-mini
Generate Text with Chat Prompt
Generate Image with Chat Prompt
streamText Multi-Step Cookbook
Markdown Chatbot with Memoization
Generate Object with File Prompt through Form Submission
Model Context Protocol (MCP) Tools
Share useChat State Across Components
Human-in-the-Loop Agent with Next.js
Render Visual Interface in Chat
Generate Text with Chat Prompt
Generate Text with Image Prompt
Generate Object with a Reasoning Model
Stream Object with Image Prompt
Record Token Usage After Streaming Object
Record Final Object after Streaming Object
Model Context Protocol (MCP) Tools
Retrieval Augmented Generation
Generate Text with Chat Prompt
Restore Messages From Database
Render Visual Interface in Chat
Stream Updates to Visual Interfaces
Record Token Usage after Streaming User Interfaces
Copy markdown
Render Visual Interface in Chat
===================================================================================================================================
We've now seen how a language model can call a function and render a component based on a conversation with the user.
When we define multiple functions in tools
, it is possible for the model to reason out the right functions to call based on whatever the user's intent is. This means that you can write a bunch of functions without the burden of implementing complex routing logic to run them.
app/page.tsx
'use client';
import { useState } from 'react';import { ClientMessage } from './actions';import { useActions, useUIState } from '@ai-sdk/rsc';import { generateId } from 'ai';
// Allow streaming responses up to 30 secondsexport const maxDuration = 30;
export default function Home() { const [input, setInput] = useState<string>(''); const [conversation, setConversation] = useUIState(); const { continueConversation } = useActions();
return ( <div> <div> {conversation.map((message: ClientMessage) => ( <div key={message.id}> {message.role}: {message.display} </div> ))} </div>
<div> <input type="text" value={input} onChange={event => { setInput(event.target.value); }} /> <button onClick={async () => { setConversation((currentConversation: ClientMessage[]) => [ ...currentConversation, { id: generateId(), role: 'user', display: input }, ]);
const message = await continueConversation(input);
setConversation((currentConversation: ClientMessage[]) => [ ...currentConversation, message, ]); }} > Send Message </button> </div> </div> );}
components/stock.tsx
export async function Stock({ symbol, numOfMonths }) { const data = await fetch( `https://api.example.com/stock/${symbol}/${numOfMonths}`, );
return ( <div> <div>{symbol}</div>
<div> {data.timeline.map(data => ( <div> <div>{data.date}</div> <div>{data.value}</div> </div> ))} </div> </div> );}
components/flight.tsx
export async function Flight({ flightNumber }) { const data = await fetch(`https://api.example.com/flight/${flightNumber}`);
return ( <div> <div>{flightNumber}</div> <div>{data.status}</div> <div>{data.source}</div> <div>{data.destination}</div> </div> );}
app/actions.tsx
'use server';
import { getMutableAIState, streamUI } from '@ai-sdk/rsc';import { openai } from '@ai-sdk/openai';import { ReactNode } from 'react';import { z } from 'zod';import { generateId } from 'ai';import { Stock } from '@/components/stock';import { Flight } from '@/components/flight';
export interface ServerMessage { role: 'user' | 'assistant'; content: string;}
export interface ClientMessage { id: string; role: 'user' | 'assistant'; display: ReactNode;}
export async function continueConversation( input: string,): Promise<ClientMessage> { 'use server';
const history = getMutableAIState();
const result = await streamUI({ model: openai('gpt-3.5-turbo'), messages: [...history.get(), { role: 'user', content: input }], text: ({ content, done }) => { if (done) { history.done((messages: ServerMessage[]) => [ ...messages, { role: 'assistant', content }, ]); }
return <div>{content}</div>; }, tools: { showStockInformation: { description: 'Get stock information for symbol for the last numOfMonths months', inputSchema: z.object({ symbol: z .string() .describe('The stock symbol to get information for'), numOfMonths: z .number() .describe('The number of months to get historical information for'), }), generate: async ({ symbol, numOfMonths }) => { history.done((messages: ServerMessage[]) => [ ...messages, { role: 'assistant', content: `Showing stock information for ${symbol}`, }, ]);
return <Stock symbol={symbol} numOfMonths={numOfMonths} />; }, }, showFlightStatus: { description: 'Get the status of a flight', inputSchema: z.object({ flightNumber: z .string() .describe('The flight number to get status for'), }), generate: async ({ flightNumber }) => { history.done((messages: ServerMessage[]) => [ ...messages, { role: 'assistant', content: `Showing flight status for ${flightNumber}`, }, ]);
return <Flight flightNumber={flightNumber} />; }, }, }, });
return { id: generateId(), role: 'assistant', display: result.value, };}
app/ai.ts
import { createAI } from '@ai-sdk/rsc';import { ServerMessage, ClientMessage, continueConversation } from './actions';
export const AI = createAI<ServerMessage[], ClientMessage[]>({ actions: { continueConversation, }, initialAIState: [], initialUIState: [],});
On this page
Render Visual Interface in Chat
Deploy and Scale AI Apps with Vercel.
Vercel delivers the infrastructure and developer experience you need to ship reliable AI-powered applications at scale.
Trusted by industry leaders: