老师您好，目前vllm0.8.5以上的版本需要在K100AI或BW系列的卡上才能正常运行，无法运行可能和您当前所使用的加速卡型号有关

pip list | grep vllm 麻烦使用这个命令查看下vllm版本，并告知下机器型号

镜像 l4-lqm-minimax-m2:vllm0.11.0-ubuntu22.04-dtk25.04.2-py3.10 root@worker-0:/public/home/u2578/gelab-zero-workspace# VLLM_USE_V1=1 python3 -m vllm.entrypoints.openai.api_server \ --model /public/home/u2578/gelab-zero-workspace/models/gelab-zero-4b/stepfun-ai/GELab-Zero-4B-preview \ --dtype float16 \ --max-model-len 4096 \ --tensor-parallel-size 2 \ --api-key gelab-zero-key-secret \ --port 8606 \ --host 0.0.0.0 \ --disable-log-stats \ --disable-log-requests INFO 12-19 04:34:18 [__init__.py:217] Automatically detected platform rocm. WARNING 12-19 04:34:24 [__init__.py:1756] argument '--disable-log-requests' is deprecated and replaced with '--enable-log-requests'. This will be removed in v0.12.0. (APIServer pid=344) INFO 12-19 04:34:24 [api_server.py:1839] vLLM API server version 0.11.0 (APIServer pid=344) INFO 12-19 04:34:24 [utils.py:233] non-default args: {'host': '0.0.0.0', 'port': 8606, 'api_key': ['gelab-zero-key-secret'], 'model': '/public/home/u2578/gelab-zero-workspace/models/gelab-zero-4b/stepfun-ai/GELab-Zero-4B-preview', 'dtype': 'float16', 'max_model_len': 4096, 'tensor_parallel_size': 2, 'disable_log_stats': True} (APIServer pid=344) INFO 12-19 04:34:24 [model.py:555] Resolved architecture: Qwen3VLForConditionalGeneration (APIServer pid=344) `torch_dtype` is deprecated! Use `dtype` instead! (APIServer pid=344) WARNING 12-19 04:34:24 [model.py:1742] Casting torch.bfloat16 to torch.float16. (APIServer pid=344) INFO 12-19 04:34:24 [model.py:1519] Using max model len 4096 (APIServer pid=344) INFO 12-19 04:34:27 [scheduler.py:205] Chunked prefill is enabled with max_num_batched_tokens=2048. (APIServer pid=344) Traceback (most recent call last): (APIServer pid=344) File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main (APIServer pid=344) return _run_code(code, main_globals, None, (APIServer pid=344) File "/usr/lib/python3.10/runpy.py", line 86, in _run_code (APIServer pid=344) exec(code, run_globals) (APIServer pid=344) File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 1953, in (APIServer pid=344) uvloop.run(run_server(args)) (APIServer pid=344) File "/usr/local/lib/python3.10/dist-packages/uvloop/__init__.py", line 69, in run (APIServer pid=344) return loop.run_until_complete(wrapper()) (APIServer pid=344) File "uvloop/loop.pyx", line 1518, in uvloop.loop.Loop.run_until_complete (APIServer pid=344) File "/usr/local/lib/python3.10/dist-packages/uvloop/__init__.py", line 48, in wrapper (APIServer pid=344) return await main (APIServer pid=344) File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 1884, in run_server (APIServer pid=344) await run_server_worker(listen_address, sock, args, **uvicorn_kwargs) (APIServer pid=344) File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 1902, in run_server_worker (APIServer pid=344) async with build_async_engine_client( (APIServer pid=344) File "/usr/lib/python3.10/contextlib.py", line 199, in __aenter__ (APIServer pid=344) return await anext(self.gen) (APIServer pid=344) File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 180, in build_async_engine_client (APIServer pid=344) async with build_async_engine_client_from_engine_args( (APIServer pid=344) File "/usr/lib/python3.10/contextlib.py", line 199, in __aenter__ (APIServer pid=344) return await anext(self.gen) (APIServer pid=344) File "/usr/local/lib/python3.10/dist-packages/vllm/entrypoints/openai/api_server.py", line 209, in build_async_engine_client_from_engine_args (APIServer pid=344) assert envs.VLLM_USE_V1 (APIServer pid=344) AssertionError

您好，请问咱使用的镜像是什么呢？麻烦贴一下具体的报错日志吧。