phoenix/litellm
1
0
Fork 0
forked from phoenix/litellm-mirror
Code Pull requests Activity

test: handle gemini overloaded model error

Browse source
This commit is contained in:
Krrish Dholakia 2024-11-14 23:47:14 +05:30
parent ffbdaf868f
commit b92700cc19
2 changed files with 8 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

9
litellm/litellm_core_utils/exception_mapping_utils.py
View file

@ -1124,10 +1124,13 @@ def exception_type( # type: ignore # noqa: PLR0915
), ),
), ),
) )
elif "500 Internal Server Error" in error_str: elif (
"500 Internal Server Error" in error_str
or "The model is overloaded." in error_str
):
exception_mapping_worked = True exception_mapping_worked = True
raise ServiceUnavailableError( raise litellm.InternalServerError(
message=f"litellm.ServiceUnavailableError: VertexAIException - {error_str}", message=f"litellm.InternalServerError: VertexAIException - {error_str}",
model=model, model=model,
llm_provider="vertex_ai", llm_provider="vertex_ai",
litellm_debug_info=extra_information, litellm_debug_info=extra_information,

4
tests/local_testing/test_stream_chunk_builder.py
View file

@ -205,8 +205,8 @@ def test_stream_chunk_builder_litellm_usage_chunks():
complete_response=True, complete_response=True,
stream_options={"include_usage": True}, stream_options={"include_usage": True},
) )
except litellm.ServiceUnavailableError as e: except litellm.InternalServerError as e:
pytest.skip(f"ServiceUnavailableError - {str(e)}") pytest.skip(f"Skipping test due to internal server error - {str(e)}")
usage: litellm.Usage = response.usage usage: litellm.Usage = response.usage