lol yeah, it’s not just you. I spent like 3 days just trying to get the dependencies right.
For debugging, I found that logging everything (like literally every step) helps a lot. Also, check out LM Studio—it’s a great tool for running local llm code interpreters without too much hassle.
As for performance, I’d say upgrade your RAM if you can. These models are memory hogs.
For debugging, I found that logging everything (like literally every step) helps a lot. Also, check out LM Studio—it’s a great tool for running local llm code interpreters without too much hassle.
As for performance, I’d say upgrade your RAM if you can. These models are memory hogs.