docker run -d \
  --name vllm-qwen35-moe \
  --gpus all \
  --ipc=host \
  -v /path/to/cache:/root/.cache \
  -p 16080:8000 \
  vllm/vllm-openai:cu130-nightly \
  --model Qwen/Qwen3.5-35B-A3B-GPTQ-Int4 \
  --served-model-name qwen35-35b-a3b \
  --quantization moe_wna16 \
  --max-num-seqs 6 \
  --max-model-len 4096 \
  --kv-cache-memory 3221225472 \
  --gpu-memory-utilization 0.50 \
  --enable-prefix-caching \
  --host 0.0.0.0 --port 8000 \
  --reasoning-parser qwen3 \
  --enable-auto-tool-choice \
  --tool-call-parser qwen3_coder \
  --language-model-only