docker run --rm --gpus '"device=1"' --name "mistral-llm1" -e CUDA_VISIBLE_DEVICES=0 -e LLAMA_CUDA_VISIBLE_DEVICES=0 -p 8001:8000 -v /opt/models/mistral:/models -t mistral-llm:cu122-avx
#docker run --rm --gpus all -p 8000:8000 -v /opt/models/mistral:/models -t mistral-llm:cu122-avx