Deploy VLLM with udocker

deploy_vllm.sh

# pip install udocker

udocker pull vllm/vllm-openai:latest

udocker create --name=vllm vllm/vllm-openai:latest

udocker setup --nvidia --force vllm

udocker run \
    --volume="/${PWD}:/workspace" \
    --env "HF_TOKEN=TOKEN" \
    vllm \
    --model "Qwen/Qwen2.5-0.5B-Instruct"
    
# you can add any arguments from: https://docs.vllm.ai/en/latest/serving/engine_args.html