Deepseek Model Deploy

apt update

apt install -y software-properties-common

add-apt-repository ppa:deadsnakes/ppa

apt install -y python3.10 python3.10-venv python3.10-dev python3.10-distutils

python3.10 -m venv deepseek-env-py310
source deepseek-env-py310/bin/activate

pip install huggingface_hub

hf download deepseek-ai/DeepSeek-R1-0528 --local-dir ./deepseek-r1-0528

apt-get install -y nvidia-container-toolkit

nvidia-ctk runtime configure --runtime=docker

systemctl restart docker

docker run -itd \
  --name vllm-gb200 \
  --gpus all \
  --ipc=host \
  --network host \
  -v /root/deepseek-r1-0528:/model/deepseek-r1-0528  \
  vllm/vllm-openai:latest \
  --model /models/deepseek-r1-0528 \
  --trust-remote-code \
  --dtype bfloat16 \
  --kv-cache-dtype fp8 \
  --tensor-parallel-size 4 \
  --gpu-memory-utilization 0.95 \
  --max-model-len 32768 \
  --max-num-seqs 128 \
  --enforce-eager --enable-auto-tool-choice  --tool-call-parser hermes

docker logs -f vllm-gb200

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "/models/DeepSeek-R1-Distill-Llama-70B",
    "messages": [
      {"role": "user", "content": "介绍一下量子计算的基本原理。"}
    ],
    "max_tokens": 100,
    "temperature": 0.7
  }'