Call Tools | AI SDK

Some models allow developers to provide a list of tools that can be called at any time during a generation. This is useful for extending the capabilities of a language model to either use logic or data to interact with systems external to the model.

http://localhost:3000

User: How is it going?

Assistant: All good, how may I help you?

What is 24 celsius in fahrenheit?

Client

Let's create a simple conversation between a user and model and place a button that will call continueConversation.

app/page.tsx

'use client';

import { useState } from 'react';
import { Message, continueConversation } from './actions';

// Allow streaming responses up to 30 seconds
export const maxDuration = 30;

export default function Home() {
  const [conversation, setConversation] = useState<Message[]>([]);
  const [input, setInput] = useState<string>('');

  return (
    <div>
      <div>
        {conversation.map((message, index) => (
          <div key={index}>
            {message.role}: {message.content}
          </div>
        ))}
      </div>

      <div>
        <input
          type="text"
          value={input}
          onChange={event => {
            setInput(event.target.value);
          }}
        />
        <button
          onClick={async () => {
            const { messages } = await continueConversation([
              ...conversation,
              { role: 'user', content: input },
            ]);

            setConversation(messages);
          }}
        >
          Send Message
        </button>
      </div>
    </div>
  );
}

Server

Now, let's implement the continueConversation action that uses generateText to generate a response to the user's question. We will use the tools parameter to specify our own function called celsiusToFahrenheit that will convert a user given value in celsius to fahrenheit.

We will use zod to specify the schema for the celsiusToFahrenheit function's parameters.

app/actions.ts

'use server';

import { generateText } from 'ai';
import { z } from 'zod';

export interface Message {
  role: 'user' | 'assistant';
  content: string;
}

export async function continueConversation(history: Message[]) {
  'use server';

  const { text, toolResults } = await generateText({
    model: 'openai/gpt-6-astra',
    instructions: 'You are a friendly assistant!',
    messages: history,
    tools: {
      celsiusToFahrenheit: {
        description: 'Converts celsius to fahrenheit',
        inputSchema: z.object({
          value: z.string().describe('The value in celsius'),
        }),
        execute: async ({ value }) => {
          const celsius = parseFloat(value);
          const fahrenheit = celsius * (9 / 5) + 32;
          return `${celsius}°C is ${fahrenheit.toFixed(2)}°F`;
        },
      },
    },
  });

  return {
    messages: [
      ...history,
      {
        role: 'assistant' as const,
        content:
          text || toolResults.map(toolResult => toolResult.result).join('\n'),
      },
    ],
  };
}