Best config to run on Apple hardware?

[pgorzelany]

Hi there! First of all, thank you for this project, you guys are doing some great work!

I am writing a MacOS and iOS app designed for Apple Silicon. How should I configure llama to have the best performance on Apple platforms? I am using the library as a Swift Package , specifically using the C api. What are the llama_model_params and llama_context_params or any other config that will give me the best performance? Or maybe the default config is good enough?

Thanks!