The prompt has 13284 tokens and the max_tokens is 4096 tokens.

[happytravelingskysheep]

Bug description

Metagpt exited halfway because of an exception due to max_tokens setting.

Bug solved method

Environment information

• LLM type and model name: yi-lightning
• System version: ubuntu 22.04
• Python version: 3.11
• MetaGPT version or branch: main
• packages version: 0.8.1
• installation method: pip

• packages version:
• installation method:

Screenshots or logs

2024-11-22 17:18:08.597 | INFO | metagpt.actions.write_code_review:run:185 - Code review and rewrite main.py: 1/2 | len(iterative_code)=12485, len(self.i_context.code_doc.content)=12485
2024-11-22 17:18:31.001 | WARNING | metagpt.utils.common:wrapper:673 - There is a exception in role's execution, in order to resume, we delete the newest role communication message in the role's memory.
2024-11-22 17:18:31.002 | ERROR | metagpt.utils.common:wrapper:655 - Exception occurs, start to serialize the project, exp:
Traceback (most recent call last):
File "/root/anaconda3/envs/metagpt/lib/python3.11/site-packages/metagpt/utils/common.py", line 650, in wrapper
result = await func(self, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/root/anaconda3/envs/metagpt/lib/python3.11/site-packages/metagpt/team.py", line 134, in run
await self.env.run()
openai.BadRequestError: Error code: 400 - {'error': {'code': 'bad_request', 'message': 'The total number of tokens in the prompt and the max_tokens must be less than or equal to 16000. The prompt has 13284 tokens and the max_tokens is 4096 tokens.', 'type': 'invalid_request_error', 'param': None}}

[voidking]

Try to set max_token in config2.yaml

llm:
  api_type: "openai"
  model: "gpt-3.5-turbo-16k" 
  base_url: "https://api.openai.com/v1" 
  api_key: "sk-xxx"
  max_token: 16000

[github-actions]

This issue has no activity in the past 30 days. Please comment on the issue if you have anything to add.