【求助】vllm部署如何支持tool参数

[TragedyN]

vllm版本: 0.6.6post1
启动命令：python -m vllm.entrypoints.openai.api_server --served-model-name Qwen2-VL-72B-Instruct --model /workspace/Qwen2-VL/models/Qwen2-VL-72B-Instruct --quantization fp8 --tensor-parallel-size 8 --port 9501
--enable-auto-tool-choice
--tool-call-parser hermes
--limit-mm-per-prompt "image=1" \

response = await self._model_client.create(
messages=messages ,
tools=tools,
extra_create_args={"tool_choice": "auto"}
)

报错："POST /v1/chat/completions HTTP/1.1" 500 Internal Server Error