能推理但是merge完部署后报错无法找到 adapter_config.json

[wookpeckerjohn]

ERROR:asyncio:Exception in callback functools.partial(<function VllmEngine.patch_remove_log..new_log_task_completion at 0x7fd8601d8d30>, error_callback=<bound method AsyncLLMEngine._error_callback of <vllm.engine.async_llm_engine.AsyncLLMEngine object at 0x7fd88c093040>>)
handle: <Handle functools.partial(<function VllmEngine.patch_remove_log..new_log_task_completion at 0x7fd8601d8d30>, error_callback=<bound method AsyncLLMEngine._error_callback of <vllm.engine.async_llm_engine.AsyncLLMEngine object at 0x7fd88c093040>>)>
Traceback (most recent call last):
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/lora/worker_manager.py", line 101, in _load_adapter
peft_helper = PEFTHelper.from_local_dir(
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/lora/peft_helper.py", line 96, in from_local_dir
with open(lora_config_path) as f:
FileNotFoundError: [Errno 2] No such file or directory: '/home/octopus/work/category/Train/ms-swift/Output/v10-20250321-154400/checkpoint-6708-merged/adapter_config.json'

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "uvloop/cbhandles.pyx", line 63, in uvloop.loop.Handle._run
File "/home/powerop/work/conda/envs/train/lib/python3.10/site-packages/swift/llm/infer/infer_engine/vllm_engine.py", line 481, in new_log_task_completion
return_value = task.result()
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/engine/async_llm_engine.py", line 825, in run_engine_loop
result = task.result()
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/engine/async_llm_engine.py", line 748, in engine_step
request_outputs = await self.engine.step_async(virtual_engine)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/engine/async_llm_engine.py", line 353, in step_async
outputs = await self.model_executor.execute_model_async(
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/executor/executor_base.py", line 250, in execute_model_async
output = await make_async(self.execute_model)(execute_model_req)
File "/home/powerop/work/conda/envs/train/lib/python3.10/concurrent/futures/thread.py", line 58, in run
result = self.fn(*self.args, **self.kwargs)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/executor/executor_base.py", line 139, in execute_model
output = self.collective_rpc("execute_model",
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/executor/uniproc_executor.py", line 56, in collective_rpc
answer = run_method(self.driver_worker, method, args, kwargs)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/utils.py", line 2196, in run_method
return func(*args, **kwargs)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/worker/worker_base.py", line 420, in execute_model
output = self.model_runner.execute_model(
File "/home/powerop/.local/lib/python3.10/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/worker/model_runner.py", line 1661, in execute_model
self.set_active_loras(model_input.lora_requests,
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/worker/model_runner.py", line 1363, in set_active_loras
self.lora_manager.set_active_adapters(lora_requests, lora_mapping)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/lora/worker_manager.py", line 165, in set_active_adapters
set_active_adapters_worker(requests, mapping, self._apply_adapters,
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/adapter_commons/utils.py", line 54, in set_active_adapters_worker
apply_adapters_func(requests)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/lora/worker_manager.py", line 225, in _apply_adapters
self.add_adapter(lora)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/lora/worker_manager.py", line 233, in add_adapter
lora = self._load_adapter(lora_request)
File "/home/powerop/.local/lib/python3.10/site-packages/vllm/lora/worker_manager.py", line 134, in _load_adapter
raise ValueError(
ValueError: Loading lora _lora failed: No adapter found for /home/octopus/work/category/Train/ms-swift/Output/v10-20250321-154400/checkpoint-6708-merged

[Jintao-Huang]

use --model instead of --adapters