采用最新版本的metagpt运行实例demo异常,异常信息为openai.types.completion_usage.CompletionUsage() argument after ** must be a mapping, not NoneType

[guyouyue]

C:\Users\glh19\MetaGPT\MetaGPT\Scripts\python.exe E:\code\python\MetaGPT\examples\di\custom_tool.py
2024-05-08 16:55:41.973 | INFO | metagpt.const:get_metagpt_package_root:29 - Package root set to E:\code\python\MetaGPT

[
    {
        "task_id": "1",
        "dependent_task_ids": [],
        "instruction": "Call the magic function with arguments 'A' and 2, then provide the result.",
        "task_type": "other"
    }
]
```2024-05-08 16:56:09.817 | WARNING  | metagpt.utils.common:wrapper:649 - There is a exception in role's execution, in order to resume, we delete the newest role communication message in the role's memory.
Traceback (most recent call last):
  File "E:\code\python\MetaGPT\metagpt\utils\common.py", line 640, in wrapper
    return await func(self, *args, **kwargs)
  File "E:\code\python\MetaGPT\metagpt\roles\role.py", line 555, in run
    rsp = await self.react()
  File "E:\code\python\MetaGPT\metagpt\roles\role.py", line 526, in react
    rsp = await self._plan_and_act()
  File "E:\code\python\MetaGPT\metagpt\roles\di\data_interpreter.py", line 95, in _plan_and_act
    raise e
  File "E:\code\python\MetaGPT\metagpt\roles\di\data_interpreter.py", line 90, in _plan_and_act
    rsp = await super()._plan_and_act()
  File "E:\code\python\MetaGPT\metagpt\roles\role.py", line 486, in _plan_and_act
    await self.planner.update_plan(goal=goal)
  File "E:\code\python\MetaGPT\metagpt\strategy\planner.py", line 75, in update_plan
    rsp = await WritePlan().run(context, max_tasks=max_tasks)
  File "E:\code\python\MetaGPT\metagpt\actions\di\write_plan.py", line 49, in run
    rsp = await self._aask(prompt)
  File "E:\code\python\MetaGPT\metagpt\actions\action.py", line 93, in _aask
    return await self.llm.aask(prompt, system_msgs)
  File "E:\code\python\MetaGPT\metagpt\provider\base_llm.py", line 150, in aask
    rsp = await self.acompletion_text(message, stream=stream, timeout=self.get_timeout(timeout))
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\_asyncio.py", line 88, in async_wrapped
    return await fn(*args, **kwargs)
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\_asyncio.py", line 47, in __call__
    do = self.iter(retry_state=retry_state)
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\__init__.py", line 314, in iter
    return fut.result()
  File "D:\Develop\python\lib\concurrent\futures\_base.py", line 451, in result
    return self.__get_result()
  File "D:\Develop\python\lib\concurrent\futures\_base.py", line 403, in __get_result
    raise self._exception
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\_asyncio.py", line 50, in __call__
    result = await fn(*args, **kwargs)
  File "E:\code\python\MetaGPT\metagpt\provider\openai_api.py", line 155, in acompletion_text
    return await self._achat_completion_stream(messages, timeout=timeout)
  File "E:\code\python\MetaGPT\metagpt\provider\openai_api.py", line 105, in _achat_completion_stream
    usage = CompletionUsage(**chunk.usage)
TypeError: openai.types.completion_usage.CompletionUsage() argument after ** must be a mapping, not NoneType

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "E:\code\python\MetaGPT\examples\di\custom_tool.py", line 36, in <module>
    asyncio.run(main())
  File "D:\Develop\python\lib\asyncio\runners.py", line 44, in run
    return loop.run_until_complete(main)
  File "D:\Develop\python\lib\asyncio\base_events.py", line 649, in run_until_complete
    return future.result()
  File "E:\code\python\MetaGPT\examples\di\custom_tool.py", line 30, in main
    await di.run("Just call the magic function with arg1 'A' and arg2 2. Tell me the result.")
  File "E:\code\python\MetaGPT\metagpt\utils\common.py", line 662, in wrapper
    raise Exception(format_trackback_info(limit=None))
Exception: Traceback (most recent call last):
  File "E:\code\python\MetaGPT\metagpt\utils\common.py", line 640, in wrapper
    return await func(self, *args, **kwargs)
  File "E:\code\python\MetaGPT\metagpt\roles\role.py", line 555, in run
    rsp = await self.react()
  File "E:\code\python\MetaGPT\metagpt\roles\role.py", line 526, in react
    rsp = await self._plan_and_act()
  File "E:\code\python\MetaGPT\metagpt\roles\di\data_interpreter.py", line 95, in _plan_and_act
    raise e
  File "E:\code\python\MetaGPT\metagpt\roles\di\data_interpreter.py", line 90, in _plan_and_act
    rsp = await super()._plan_and_act()
  File "E:\code\python\MetaGPT\metagpt\roles\role.py", line 486, in _plan_and_act
    await self.planner.update_plan(goal=goal)
  File "E:\code\python\MetaGPT\metagpt\strategy\planner.py", line 75, in update_plan
    rsp = await WritePlan().run(context, max_tasks=max_tasks)
  File "E:\code\python\MetaGPT\metagpt\actions\di\write_plan.py", line 49, in run
    rsp = await self._aask(prompt)
  File "E:\code\python\MetaGPT\metagpt\actions\action.py", line 93, in _aask
    return await self.llm.aask(prompt, system_msgs)
  File "E:\code\python\MetaGPT\metagpt\provider\base_llm.py", line 150, in aask
    rsp = await self.acompletion_text(message, stream=stream, timeout=self.get_timeout(timeout))
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\_asyncio.py", line 88, in async_wrapped
    return await fn(*args, **kwargs)
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\_asyncio.py", line 47, in __call__
    do = self.iter(retry_state=retry_state)
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\__init__.py", line 314, in iter
    return fut.result()
  File "D:\Develop\python\lib\concurrent\futures\_base.py", line 451, in result
    return self.__get_result()
  File "D:\Develop\python\lib\concurrent\futures\_base.py", line 403, in __get_result
    raise self._exception
  File "C:\Users\glh19\MetaGPT\MetaGPT\lib\site-packages\tenacity\_asyncio.py", line 50, in __call__
    result = await fn(*args, **kwargs)
  File "E:\code\python\MetaGPT\metagpt\provider\openai_api.py", line 155, in acompletion_text
    return await self._achat_completion_stream(messages, timeout=timeout)
  File "E:\code\python\MetaGPT\metagpt\provider\openai_api.py", line 105, in _achat_completion_stream
    usage = CompletionUsage(**chunk.usage)
TypeError: openai.types.completion_usage.CompletionUsage() argument after ** must be a mapping, not NoneType


进程已结束，退出代码为 1

[guyouyue]

目前分析是在
async def _achat_completion_stream(self, messages: list[dict], timeout=USE_CONFIG_TIMEOUT) -> str:
response: AsyncStream[ChatCompletionChunk] = await self.aclient.chat.completions.create(
**self._cons_kwargs(messages, timeout=self.get_timeout(timeout)), stream=True
)
usage = None
collected_messages = []
async for chunk in response:
chunk_message = chunk.choices[0].delta.content or "" if chunk.choices else "" # extract the message
finish_reason = (
chunk.choices[0].finish_reason if chunk.choices and hasattr(chunk.choices[0], "finish_reason") else None
)
log_llm_stream(chunk_message)
collected_messages.append(chunk_message)
if finish_reason:
if hasattr(chunk, "usage"):
# Some services have usage as an attribute of the chunk, such as Fireworks
usage = CompletionUsage(**chunk.usage)
elif hasattr(chunk.choices[0], "usage") :
# The usage of some services is an attribute of chunk.choices[0], such as Moonshot
usage = CompletionUsage(**chunk.choices[0].usage)
elif "openrouter.ai" in self.config.base_url:
# due to it get token cost from api
usage = await get_openrouter_tokens(chunk)
中的if hasattr(chunk, "usage")和elif hasattr(chunk.choices[0], "usage")没有判空导致的

[aa875982361]

i have a some problem

[huangpan2507]

yes, I have the same problem with metagpt version=0.8.1, also happend again after i use pip install metagpt==0.7.1

[nzk1912]

me too. bug report #1250

[CHENZHEN3078]

yeah, I have the same problem

C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\python.exe C:\Users\NEZ1SGH\metagpt\MetaGPT\examples\hello_world.py
2024-05-10 16:33:10.087 | INFO | metagpt.const:get_metagpt_package_root:29 - Package root set to C:\Users\NEZ1SGH\metagpt\MetaGPT
2024-05-10 16:33:16.477 | INFO | main:ask_and_print:15 - Q: what's your name?
I am an AI assistant and you can call me Assistant. How can I help you today?Traceback (most recent call last):
File "C:\Users\NEZ1SGH\metagpt\MetaGPT\examples\hello_world.py", line 45, in
asyncio.run(main())
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\asyncio\runners.py", line 44, in run
return loop.run_until_complete(main)
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\asyncio\base_events.py", line 647, in run_until_complete
return future.result()
File "C:\Users\NEZ1SGH\metagpt\MetaGPT\examples\hello_world.py", line 39, in main
await ask_and_print("what's your name?", llm, "I'm a helpful AI assistant.")
File "C:\Users\NEZ1SGH\metagpt\MetaGPT\examples\hello_world.py", line 16, in ask_and_print
rsp = await llm.aask(question, system_msgs=[system_prompt])
File "C:\Users\NEZ1SGH\metagpt\MetaGPT\metagpt\provider\base_llm.py", line 150, in aask
rsp = await self.acompletion_text(message, stream=stream, timeout=self.get_timeout(timeout))
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\site-packages\tenacity_asyncio.py", line 88, in async_wrapped
return await fn(*args, **kwargs)
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\site-packages\tenacity_asyncio.py", line 47, in call
do = self.iter(retry_state=retry_state)
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\site-packages\tenacity_init_.py", line 314, in iter
return fut.result()
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\concurrent\futures_base.py", line 439, in result
return self.__get_result()
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\concurrent\futures_base.py", line 391, in __get_result
raise self._exception
File "C:\Users\NEZ1SGH\AppData\Local\anaconda3\envs\myenv\lib\site-packages\tenacity_asyncio.py", line 50, in call
result = await fn(*args, **kwargs)
File "C:\Users\NEZ1SGH\metagpt\MetaGPT\metagpt\provider\openai_api.py", line 155, in acompletion_text
return await self._achat_completion_stream(messages, timeout=timeout)
File "C:\Users\NEZ1SGH\metagpt\MetaGPT\metagpt\provider\openai_api.py", line 105, in _achat_completion_stream
usage = CompletionUsage(**chunk.usage)
TypeError: openai.types.completion_usage.CompletionUsage() argument after ** must be a mapping, not NoneType

[hs3180]

Add a None filter at this line can avoid this exception.

if hasattr(chunk, "usage") and chunk.usage is not None:

MetaGPT/metagpt/provider/openai_api.py

Line 103 in f201b2f

if hasattr(chunk, "usage"):

[baixiaolu]

我的临时解决办法：
MetaGPT/metagpt/provider/openai_api.py文件102行左右和usage计算相关的代码注释掉。共3块代码：

1. 102-111行：if finish_reason:开头的整块代码，全部注释。
2. 115-117行：if not usage:开头的整块代码，全部注释。
3. 119行：self._update_costs(usage)，注释1行。
   虽然失去了计算usage的功能，但代码可正常运行了。

注意，如果很依赖usage的计算，慎重采用本方法。

[chg0901]

Add a None filter at this line can avoid this exception.

if hasattr(chunk, "usage") and chunk.usage is not None:

MetaGPT/metagpt/provider/openai_api.py

Line 103 in f201b2f

if hasattr(chunk, "usage"):

One of my friend use this methods and it worked but not for me.

[chg0901]

我的临时解决办法： MetaGPT/metagpt/provider/openai_api.py文件102行左右和usage计算相关的代码注释掉。共3块代码：

1. 102-111行：if finish_reason:开头的整块代码，全部注释。
2. 115-117行：if not usage:开头的整块代码，全部注释。
3. 119行：self._update_costs(usage)，注释1行。
   虽然失去了计算usage的功能，但代码可正常运行了。

注意，如果很依赖usage的计算，慎重采用本方法。

This works for me, I use the api of DeepSeek, a way similar to the api of openai

llm:
  api_type: "openai" 
  model: "deepseek-chat"  
  base_url: "https://api.deepseek.com" 
  api_key: ''

[garylin2099]

Hey guys, thanks for reporting the issue. For the latest main branch, it should be fixed in #1253 . For pypi package metagpt==0.8.1, try using openai==1.6.1 as indicated in the requirement.txt, the issue is probably due to the openai package upgrade. Please check if this resolves the issue.