Markhit

/

CodeLlama3-8B-Python

like 13

Not yet, I am currently evaluating my model.The evaluation results will be published soon.

not so good, output, strugggles to get the context, i got better results with llama 3 8b 262k instruct model gguf, and codeqwn 7b gguf model, both Q8s, llama.cpp used, on laptop with 16gb ram, no dedicated gpu, onlu cpu inference