From the course: The AI Ecosystem for Developers: Models, Datasets, and APIs

Unlock this course with a free trial

Join today to access over 24,900 courses taught by industry experts.

Generate code and architecture with GitHub Models

Generate code and architecture with GitHub Models

From the course: The AI Ecosystem for Developers: Models, Datasets, and APIs

Start my 1-month free trial Buy for my team

Generate code and architecture with GitHub Models

“

- [Instructor] By now we've explored enough models to understand that the output of AI models largely depend on their primary training data. In this video, we'll compare two models from Mistral, one optimized for code, and another for general purpose. We'll be looking at Codestral 25.01 and Mistral Large 24.11. I should mention again, as of the time you're exploring these, there could be other variants that are not those ones that we are testing. It's a great exercise for you to poke around and figure it out. To assess the models, remember to go to github.com/marketplace/models. And on the select model, let's look for Codestral 25.01. And then we're going to enter this prompt. "What will the code architecture for a chatbot software using an AI API look like?" Okay, so Codestral is going into the details of what the high-level architecture of this type of application, an AI application, that it's showing what the things designed in the chatbot software, what the architecture will look…

Contents