> "write a function that does x" Which model is optimized to do that? This is what I want out of LLMs! And also talking high level architecture (without any co

> "write a function that does x" Which model is optimized to do that? This is what I want out of LLMs! And also talking high level architecture (without any code) and library discovery, but I guess the general talking models are good for that...