services:
  vllm-eval:
    image: vllm/vllm-openai:cu130-nightly-aarch64
    container_name: vllm-eval
    restart: "no"
    ipc: host
    ports:
      - "8001:8000"
    volumes:
      - /home/aleleba/ft-models/Qwen3.6-35B-A3B-mcp-bf16:/model:ro
    command:
      - "--model=/model"
      - "--served-model-name=qwen3.6-35b-a3b-mcp-bf16"
      - "--tensor-parallel-size=1"
      - "--max-model-len=32768"
      - "--enable-auto-tool-choice"
      - "--tool-call-parser=qwen3_coder"
      - "--reasoning-parser=qwen3"
      - "--default-chat-template-kwargs={\"preserve_thinking\": true}"
      - "--trust-remote-code"
    deploy:
      resources:
        reservations:
          devices:
            - driver: nvidia
              count: 1
              capabilities: [gpu]