Sophisticated docker builds for parent projects oobabooga/GPTQ-for-Llama and qwopqwop200/GPTQ-for-Llama
Easy setup. Compatible. Tweakable. Scaleable.
source alias.gptq-for-llama
gptq-for-llama convert_llama_weights_to_hf.py --input_dir /path/to/downloaded/llama/weights --model_size 7B --output_dir ./llama-hf