Using Ollama's Text Generation Interface

00:00 In the last two lessons, you used Ollama’s chat() function. But if you don’t need to have a conversation or understand context, then the generate() function is great for one-shot prompts.

00:13 Here’s how to use it. Start off by importing the generate() function. So from ollama import generate. Instead of having a list of messages, you can just have a single prompt string.

00:28 So prompt equals a string, "Describe Python in one sentence."

00:36 Next, you’ll call the generate() function, and it’s going to return a response. So response = generate, call it, and pass in a model.

00:47 You can use llama3.2, like last lesson, and prompt = prompt. So you’re just passing in a string instead of a list of messages. And now you can print response.response

01:05 and run it to make sure that it works.

01:10 So you can see that this is a simpler interface. You’re not dealing with messages or roles, and this function is convenient for performing one-off tasks like summarizing, rewriting, and rephrasing text.

01:25 In the next lesson, you’ll see how to use it to generate Python code.