services:
  vllm:
    image: vllm/vllm-openai:latest
    container_name: vllm
    restart: unless-stopped
    ipc: host # vLLM braucht shared memory für Tensor-Parallelism

    ports:
      - "8000:8000" # OpenAI-kompatible API

    volumes:
      - /opt/vLLM/models:/root/.cache/huggingface # HF-Model-Cache
      - /opt/vLLM/config:/vllm-workspace/config

    environment:
      - NVIDIA_VISIBLE_DEVICES=all
      - NVIDIA_DRIVER_CAPABILITIES=compute,utility

    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: all
              capabilities: [gpu]

    command: >
      --model Qwen/Qwen2.5-14B-Instruct-AWQ
      --host 0.0.0.0
      --port 8000
      --gpu-memory-utilization 0.90
      --max-model-len 16384
      --enable-auto-tool-choice
      --tool-call-parser hermes