Let's look at the most basic functionality of the AI SDK, generating some text. You need two inputs to generate some text. You need:
I've given you a couple of to-dos inside the main.ts file. The first to-do is to choose a model and then instantiate it. This means going up to the @ai-sdk/google import here, grabbing the google function and then calling it with the model that you want to choose.
// Import the necessary functionsimport { google } from '@ai-sdk/google';import { generateText } from 'ai';// TODO: Choose a model. I recommend using the Google Gemini model:// gemini-2.0-flashconst model = TODO;
We've got a prompt here asking what is the capital of France:
const prompt = 'What is the capital of France?';
And then this result is going to be the result of this generateTextcall that we get from the ai package here. You're going to call generateText, passing in the model that you chose and passing in the prompt and then you're going to await it and get the result back.
const result = TODO; // TODO: Use generateText to get the result
Finally, we're going to console.log the text:
console.log(result.text);
This means that when you run this exercise, you should see the LLM that we contacted answering the question that we asked it.
That's it, a nice, simple exercise to start with.
Good luck and I will see you in the solution.
Choose a model by replacing the TODO in the model declaration with the Google Gemini model.
Use the generateText function to get the result by replacing the TODO in the result declaration.
Run the code and check the terminal output to verify that you get a response about the capital of France.
If you get stuck, check the solution.
