From the course: Creating AI Applications with Python and GitHub Models

Unlock this course with a free trial

Join today to access over 25,500 courses taught by industry experts.

Solution: Choose your LLM

Solution: Choose your LLM

From the course: Creating AI Applications with Python and GitHub Models

Start my 1-month free trial Buy for my team

Solution: Choose your LLM

“

(upbeat music) - [Instructor] So here's one way to go about things. So let's head over to github.com/marketplace. I'm going to head over to Models. And because I'm looking for a smaller model, I'm going to go ahead and select Low latency. Now, one of the things that I like to do is to have a baseline in order for me to be able to compare models. So my baseline for smaller models is typically one of the ones from OpenAI. So I'm going to use GPT-4o mini. And let's go ahead and try the Phi-3-mini instruct model. Okay, so I'm going to head over to Playground. And now let me add the GPT-4o mini model, and we are going to use that as our comparison. Now let's see how these models do with a couple of jokes. So let's try our first one. So explain this joke. And let's see how both these models do. So go ahead and pause the screen and decide which of these two models you prefer. So I definitely prefer the one from the OpenAI GPT-4o mini model. Right, I think it's shorter and to the point. Now…

Contents