vllm serve /root/data/qwen3.5-35b-a3b --tensor-parallel-size 8 --trust-remote-code --max-model-len 32768 --host 0.0.0.0 --port 8000 --api-key root@8301 --gpu-memory-utilization 0.9 --served-model-name qwen3.5-35b-a3b