This repository was archived by the owner on Dec 16, 2025. It is now read-only.
Description Description of the bug:
Description
I'm consistently encountering an Internal Server Error (HTTP 500) when trying to use the Google Gemini API through the Inspect evaluation framework. This error occurs during the generate_content call. I'm currently on the free trial.
Steps to Reproduce
Set up an evaluation using the Inspect framework
Configure the evaluation to use the Gemini model (in my case, google/gemini-1.5-pro)
Run the evaluation
Error Message
InternalServerError: 500 An internal error has occurred. Please retry or report in https://developers.generativeai.google/guide/troubleshooting
Environment
Operating System: MacOS Sonoma 14.5
Python version: 3.12.4
Package Versions
google-ai-generativelanguage: 0.6.6
google-api-core: 2.19.2
google-api-python-client: 2.143.0
google-auth: 2.34.0
google-auth-httplib2: 0.2.0
google-generativeai: 0.7.2
googleapis-common-protos: 1.65.0
Full error traceback:
╭─ benchmarks/gpqa (78 samples): google/gemini-1.5-pro ─────────────────────────────────────────────────────────────────────╮
│ ╭─────────────────────────────── Traceback (most recent call last) ────────────────────────────────╮ dataset: (samples) │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ scorer: choice │
│ │ in task_run │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in task_run_sample │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in __call__ │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in task_run_sample │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in solve │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in generate │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in task_generate │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in generate │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in _generate │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/tenacity/as… │ │
│ │ in async_wrapped │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/tenacity/as… │ │
│ │ in __call__ │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/tenacity/as… │ │
│ │ in iter │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/tenacity/_u… │ │
│ │ in inner │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/tenacity/__… │ │
│ │ in <lambda> │ │
│ │ │ │
│ │ /opt/homebrew/Cellar/python@3.12/3.12.4/Frameworks/Python.framework/Versions/3.12/lib/python3.1… │ │
│ │ in result │ │
│ │ │ │
│ │ 446 │ │ │ │ if self._state in [CANCELLED, CANCELLED_AND_NOTIFIED]: │ │
│ │ 447 │ │ │ │ │ raise CancelledError() │ │
│ │ 448 │ │ │ │ elif self._state == FINISHED: │ │
│ │ ❱ 449 │ │ │ │ │ return self.__get_result() │ │
│ │ 450 │ │ │ │ │ │
│ │ 451 │ │ │ │ self._condition.wait(timeout) │ │
│ │ 452 │ │
│ │ │ │
│ │ /opt/homebrew/Cellar/python@3.12/3.12.4/Frameworks/Python.framework/Versions/3.12/lib/python3.1… │ │
│ │ in __get_result │ │
│ │ │ │
│ │ 398 │ def __get_result(self): │ │
│ │ 399 │ │ if self._exception: │ │
│ │ 400 │ │ │ try: │ │
│ │ ❱ 401 │ │ │ │ raise self._exception │ │
│ │ 402 │ │ │ finally: │ │
│ │ 403 │ │ │ │ # Break a reference cycle with the exception in self._exception │ │
│ │ 404 │ │ │ │ self = None │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/tenacity/as… │ │
│ │ in __call__ │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in generate │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/inspect_ai/… │ │
│ │ in generate │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/google/gene… │ │
│ │ in generate_content_async │ │
│ │ │ │
│ │ 382 │ │ │ │ │ ) │ │
│ │ 383 │ │ │ │ return await generation_types.AsyncGenerateContentResponse.from_aiterato │ │
│ │ 384 │ │ │ else: │ │
│ │ ❱ 385 │ │ │ │ response = await self._async_client.generate_content( │ │
│ │ 386 │ │ │ │ │ request, │ │
│ │ 387 │ │ │ │ │ **request_options, │ │
│ │ 388 │ │ │ │ ) │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/google/ai/g… │ │
│ │ in generate_content │ │
│ │ │ │
│ │ 403 │ │ self._client._validate_universe_domain() │ │
│ │ 404 │ │ │ │
│ │ 405 │ │ # Send the request. │ │
│ │ ❱ 406 │ │ response = await rpc( │ │
│ │ 407 │ │ │ request, │ │
│ │ 408 │ │ │ retry=retry, │ │
│ │ 409 │ │ │ timeout=timeout, │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/google/api_… │ │
│ │ in retry_wrapped_func │ │
│ │ │ │
│ │ 227 │ │ │ sleep_generator = exponential_sleep_generator( │ │
│ │ 228 │ │ │ │ self._initial, self._maximum, multiplier=self._multiplier │ │
│ │ 229 │ │ │ ) │ │
│ │ ❱ 230 │ │ │ return await retry_target( │ │
│ │ 231 │ │ │ │ functools.partial(func, *args, **kwargs), │ │
│ │ 232 │ │ │ │ predicate=self._predicate, │ │
│ │ 233 │ │ │ │ sleep_generator=sleep_generator, │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/google/api_… │ │
│ │ in retry_target │ │
│ │ │ │
│ │ 157 │ │ # This function explicitly must deal with broad exceptions. │ │
│ │ 158 │ │ except Exception as exc: │ │
│ │ 159 │ │ │ # defer to shared logic for handling errors │ │
│ │ ❱ 160 │ │ │ _retry_error_helper( │ │
│ │ 161 │ │ │ │ exc, │ │
│ │ 162 │ │ │ │ deadline, │ │
│ │ 163 │ │ │ │ sleep, │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/google/api_… │ │
│ │ in _retry_error_helper │ │
│ │ │ │
│ │ 209 │ │ │ RetryFailureReason.NON_RETRYABLE_ERROR, │ │
│ │ 210 │ │ │ original_timeout, │ │
│ │ 211 │ │ ) │ │
│ │ ❱ 212 │ │ raise final_exc from source_exc │ │
│ │ 213 │ if on_error_fn is not None: │ │
│ │ 214 │ │ on_error_fn(exc) │ │
│ │ 215 │ if deadline is not None and time.monotonic() + next_sleep > deadline: │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/google/api_… │ │
│ │ in retry_target │ │
│ │ │ │
│ │ 152 │ │ │
│ │ 153 │ for sleep in sleep_generator: │ │
│ │ 154 │ │ try: │ │
│ │ ❱ 155 │ │ │ return await target() │ │
│ │ 156 │ │ # pylint: disable=broad-except │ │
│ │ 157 │ │ # This function explicitly must deal with broad exceptions. │ │
│ │ 158 │ │ except Exception as exc: │ │
│ │ │ │
│ │ /Users/lenni/Documents/GitHub/biology-benchmarks/.venv/lib/python3.12/site-packages/google/api_… │ │
│ │ in __await__ │ │
│ │ │ │
│ │ 85 │ │ │ response = yield from self._call.__await__() │ │
│ │ 86 │ │ │ return response │ │
│ │ 87 │ │ except grpc.RpcError as rpc_error: │ │
│ │ ❱ 88 │ │ │ raise exceptions.from_grpc_error(rpc_error) from rpc_error │ │
│ │ 89 │ │
│ │ 90 │ │
│ │ 91 class _WrappedStreamResponseMixin(Generic[P], _WrappedCall): │ │
│ ╰──────────────────────────────────────────────────────────────────────────────────────────────────╯ │
│ InternalServerError: 500 An internal error has occurred. Please retry or report in │
│ https://developers.generativeai.google/guide/troubleshooting
Actual vs expected behavior:
No response
Any other information you'd like to share?
No response
Reactions are currently unavailable
Description of the bug:
Description
I'm consistently encountering an Internal Server Error (HTTP 500) when trying to use the Google Gemini API through the Inspect evaluation framework. This error occurs during the
generate_contentcall. I'm currently on the free trial.Steps to Reproduce
google/gemini-1.5-pro)Error Message
InternalServerError: 500 An internal error has occurred. Please retry or report in https://developers.generativeai.google/guide/troubleshooting
Environment
Package Versions
Full error traceback:
Actual vs expected behavior:
No response
Any other information you'd like to share?
No response