Comment by haellsigh

I added it in my oh-my-pi configuration before (it's OpenAI compatible), but Llama 3 8B is just absolutely unusable for anything coding related. It is very fast and the latency is very good however.