Skip to content

Regression in c980000 #219

@pjcreath

Description

@pjcreath

Issue

My first query gets sent and produces the right response:

LiteLLM API Error: litellm.InternalServerError: litellm.InternalServerError: VertexAIException - {
  "error": {
    "code": 503,
    "message": "The model is overloaded. Please try again later.",
    "status": "UNAVAILABLE"
  }
}

Empty response received from LLM. Check your provider account?

But all subsequent attempts to retry produce the following:

Traceback (most recent call last):
  File "aider-ce/lib/python3.12/site-packages/aider/coders/base_coder.py", line 2175, in send_message
    async for chunk in self.send(messages, tools=self.get_tool_list()):
  File "aider-ce/lib/python3.12/site-packages/aider/coders/base_coder.py", line 2907, in send
    hash_object, completion = await model.send_completion(
                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "aider-ce/lib/python3.12/site-packages/aider/models.py", line 913, in send_completion
    messages = model_request_parser(self, messages)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "aider-ce/lib/python3.12/site-packages/aider/helpers/requests.py", line 40, in model_request_parser
    messages = thought_signature(model, messages)
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "aider-ce/lib/python3.12/site-packages/aider/helpers/requests.py", line 10, in thought_signature
    for call in tool_calls:
                ^^^^^^^^^^
TypeError: 'NoneType' object is not iterable
'NoneType' object is not iterable

Prior to this commit, I would instead receive the following:

LiteLLM API Error: litellm.RateLimitError: litellm.RateLimitError: geminiException - {
  "error": {
    "code": 429,
    "message": "You exceeded your current quota, please check your plan and billing details. For more information on this error, head to: https://ai.google.dev/gemini-api/docs/rate-limits. To monitor your current usage, head to: https://ai.dev/usage?tab=rate-limit. \n* Quota exceeded for metric: generativelanguage.googleapis.com/generate_content_free_tier_input_token_count, limit: 125000, model: gemini-2.5-pro\nPlease retry in 22.492951154s.",
    "status": "RESOURCE_EXHAUSTED",
    "details": [
      {
        "@type": "type.googleapis.com/google.rpc.Help",
        "links": [
          {
            "description": "Learn more about Gemini API quotas",
            "url": "https://ai.google.dev/gemini-api/docs/rate-limits"
          }
        ]
      },
      {
        "@type": "type.googleapis.com/google.rpc.QuotaFailure",
        "violations": [
          {
            "quotaMetric": "generativelanguage.googleapis.com/generate_content_free_tier_input_token_count",
            "quotaId": "GenerateContentInputTokensPerModelPerMinute-FreeTier",
            "quotaDimensions": {
              "location": "global",
              "model": "gemini-2.5-pro"
            },
            "quotaValue": "125000"
          }
        ]
      },
      {
        "@type": "type.googleapis.com/google.rpc.RetryInfo",
        "retryDelay": "22s"
      }
    ]
  }
}

Empty response received from LLM. Check your provider account?

I don't yet know whether the error is specifically related to the 503 overload or the quota failure, as I haven't been able to get farther.

Version and model info

Aider-CE v0.88.37
Main model: gemini/gemini-2.5-pro with ask edit format
Weak model: gemini/gemini-2.5-flash
Git repo: .git with 7,338 files
Repo-map: using 4096 tokens, files refresh

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions