使用昇腾npu训练拓展模型，出现报错

[xiaoyaoyouh]

Traceback (most recent call last):
File "/media/t1/gjq123/verl/verl/trainer/main_ppo.py", line 45, in main
run_ppo(config)
File "/media/t1/gjq123/verl/verl/trainer/main_ppo.py", line 99, in run_ppo
ray.get(runner.run.remote(config))
File "/root/miniconda3/envs/MindSpeed/lib/python3.10/site-packages/ray/_private/auto_init_hook.py", line 22, in auto_init_wrapper
return fn(*args, **kwargs)
File "/root/miniconda3/envs/MindSpeed/lib/python3.10/site-packages/ray/_private/client_mode_hook.py", line 104, in wrapper
return func(*args, **kwargs)
File "/root/miniconda3/envs/MindSpeed/lib/python3.10/site-packages/ray/_private/worker.py", line 2967, in get
values, debugger_breakpoint = worker.get_objects(
File "/root/miniconda3/envs/MindSpeed/lib/python3.10/site-packages/ray/_private/worker.py", line 1015, in get_objects
raise value.as_instanceof_cause()
ray.exceptions.RayTaskError(ImportError): ray::TaskRunner.run() (pid=1273709, ip=192.168.0.140, actor_id=73e484a0a654932a78182b6301000000, repr=<main_ppo.TaskRunner object at 0xffcf91af5b40>)
File "/media/t1/gjq123/verl/verl/trainer/main_ppo.py", line 359, in run
trainer.init_workers()
File "/media/t1/gjq123/verl/verl/trainer/ppo/ray_trainer.py", line 958, in init_workers
self.actor_rollout_wg.init_model()
File "/media/t1/gjq123/verl/verl/single_controller/ray/base.py", line 54, in call
output = ray.get(output)
ray.exceptions.RayTaskError(ImportError): ray::WorkerDict.actor_rollout_init_model() (pid=1275501, ip=192.168.0.140, actor_id=e2ebf9b29349f81297021be401000000, repr=<verl.single_controller.ray.base.WorkerDict object at 0xffcfb5bc0d90>)
File "/root/miniconda3/envs/MindSpeed/lib/python3.10/concurrent/futures/_base.py", line 458, in result
return self.__get_result()
File "/root/miniconda3/envs/MindSpeed/lib/python3.10/concurrent/futures/_base.py", line 403, in __get_result
raise self._exception
File "/media/t1/gjq123/verl/verl/single_controller/ray/base.py", line 844, in func
return getattr(self.worker_dict[key], name)(*args, **kwargs)
File "/media/t1/gjq123/verl/verl/single_controller/base/decorator.py", line 462, in inner
return func(*args, **kwargs)
File "/media/t1/gjq123/verl/verl/utils/transferqueue_utils.py", line 314, in dummy_inner
output = func(*args, **kwargs)
File "/media/t1/gjq123/verl/verl/workers/fsdp_workers.py", line 847, in init_model
self._build_rollout(trust_remote_code=self.config.model.get("trust_remote_code", False))
File "/media/t1/gjq123/verl/verl/workers/fsdp_workers.py", line 650, in _build_rollout
self.rollout = get_rollout_class(rollout_config.name, rollout_config.mode)(
File "/media/t1/gjq123/verl/verl/workers/rollout/base.py", line 100, in get_rollout_class
rollout_module = importlib.import_module(module_name)
File "/root/miniconda3/envs/MindSpeed/lib/python3.10/importlib/init.py", line 126, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
File "", line 1050, in _gcd_import
File "", line 1027, in _find_and_load
File "", line 1006, in _find_and_load_unlocked
File "", line 688, in _load_unlocked
File "", line 883, in exec_module
File "", line 241, in _call_with_frames_removed
File "/media/t1/gjq123/verl/verl/workers/rollout/vllm_rollout/init.py", line 17, in
from .vllm_rollout import vLLMAsyncRollout # noqa: F401
File "/media/t1/gjq123/verl/verl/workers/rollout/vllm_rollout/vllm_rollout.py", line 57, in
from verl.third_party.vllm import VLLM_SLEEP_LEVEL, get_version
File "/media/t1/gjq123/verl/verl/third_party/vllm/init.py", line 44, in
from vllm import LLM
ImportError: cannot import name 'LLM' from 'vllm' (unknown location)

[xiaoyaoyouh]

相关配置：
vllm 0.7.3+empty /media/t1/gjq123/vllm
vllm_ascend 0.7.3.post2.dev0+gc69ceaca1.d20260112

[tardis-key]

https://quay.io/repository/ascend/verl?tab=tags&tag=latest
感觉就是环境问题，直接用社区的每日镜像把