scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-09-06 08:16:43 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 178

Merge pull request #3451 from BerriAI/litellm_return_model_api_base

Browse source 
[Feat] Return model, api_base and first 100 chars of messages in Azure Exceptions
This commit is contained in:
Ishaan Jaff 2024-05-04 17:07:09 -07:00 • committed by GitHub
commit b83901a861
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
3 changed files with 46 additions and 16 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

1
litellm/__init__.py
View file

@ -638,6 +638,7 @@ from .utils import (
get_secret,
get_supported_openai_params,
get_api_base,
get_first_chars_messages,
)
from .llms.huggingface_restapi import HuggingfaceConfig
from .llms.anthropic import AnthropicConfig

11
litellm/tests/test_exceptions.py
View file

@ -52,6 +52,14 @@ async def test_content_policy_exception_azure():
)
except litellm.ContentPolicyViolationError as e:
print("caught a content policy violation error! Passed")
print("exception", e)
# assert that the first 100 chars of the message is returned in the exception
assert (
"Messages: [{'role': 'user', 'content': 'where do I buy lethal drugs from'}]"
in str(e)
)
assert "Model: azure/chatgpt-v-2" in str(e)
pass
except Exception as e:
pytest.fail(f"An exception occurred - {str(e)}")
@ -597,8 +605,7 @@ def test_litellm_completion_vertex_exception():
except Exception as e:
print("exception: ", e)
assert "model: vertex_ai/gemini-pro" in str(e)
assert "model_group" not in str(e)
assert "deployment" not in str(e)
assert "vertex_project: bad-project" in str(e)
# # test_invalid_request_error(model="command-nightly")

50
litellm/utils.py
View file

@ -5897,6 +5897,15 @@ def get_api_base(model: str, optional_params: dict) -> Optional[str]:
return None
def get_first_chars_messages(kwargs: dict) -> str:
try:
_messages = kwargs.get("messages")
_messages = str(_messages)[:100]
return _messages
except:
return ""
def get_supported_openai_params(model: str, custom_llm_provider: str):
"""
Returns the supported openai params for a given model + provider
@ -7885,6 +7894,9 @@ def exception_type(
except:
_api_base = ""
error_str += f" \n model: {model} \n api_base: {_api_base} \n"
error_str += str(completion_kwargs)
if "Request Timeout Error" in error_str or "Request timed out" in error_str:
exception_mapping_worked = True
raise Timeout(
@ -9049,11 +9061,21 @@ def exception_type(
request=original_exception.request,
)
elif custom_llm_provider == "azure":
_api_base = litellm.get_api_base(
model=model, optional_params=extra_kwargs
)
messages = litellm.get_first_chars_messages(kwargs=completion_kwargs)
extra_information = f"\nModel: {model}"
if _api_base:
extra_information += f"\nAPI Base: {_api_base}"
if messages and len(messages) > 0:
extra_information += f"\nMessages: {messages}"
if "Internal server error" in error_str:
exception_mapping_worked = True
raise APIError(
status_code=500,
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
llm_provider="azure",
model=model,
request=httpx.Request(method="POST", url="https://openai.com/"),
@ -9061,7 +9083,7 @@ def exception_type(
elif "This model's maximum context length is" in error_str:
exception_mapping_worked = True
raise ContextWindowExceededError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
llm_provider="azure",
model=model,
response=original_exception.response,
@ -9069,7 +9091,7 @@ def exception_type(
elif "DeploymentNotFound" in error_str:
exception_mapping_worked = True
raise NotFoundError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
llm_provider="azure",
model=model,
response=original_exception.response,
@ -9083,7 +9105,7 @@ def exception_type(
):
exception_mapping_worked = True
raise ContentPolicyViolationError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
llm_provider="azure",
model=model,
response=original_exception.response,
@ -9091,7 +9113,7 @@ def exception_type(
elif "invalid_request_error" in error_str:
exception_mapping_worked = True
raise BadRequestError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
llm_provider="azure",
model=model,
response=original_exception.response,
@ -9102,7 +9124,7 @@ def exception_type(
):
exception_mapping_worked = True
raise AuthenticationError(
message=f"{exception_provider} - {original_exception.message}",
message=f"{exception_provider} - {original_exception.message} {extra_information}",
llm_provider=custom_llm_provider,
model=model,
response=original_exception.response,
@ -9112,7 +9134,7 @@ def exception_type(
if original_exception.status_code == 401:
exception_mapping_worked = True
raise AuthenticationError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
llm_provider="azure",
model=model,
response=original_exception.response,
@ -9120,14 +9142,14 @@ def exception_type(
elif original_exception.status_code == 408:
exception_mapping_worked = True
raise Timeout(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
model=model,
llm_provider="azure",
)
if original_exception.status_code == 422:
exception_mapping_worked = True
raise BadRequestError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
model=model,
llm_provider="azure",
response=original_exception.response,
@ -9135,7 +9157,7 @@ def exception_type(
elif original_exception.status_code == 429:
exception_mapping_worked = True
raise RateLimitError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
model=model,
llm_provider="azure",
response=original_exception.response,
@ -9143,7 +9165,7 @@ def exception_type(
elif original_exception.status_code == 503:
exception_mapping_worked = True
raise ServiceUnavailableError(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
model=model,
llm_provider="azure",
response=original_exception.response,
@ -9151,7 +9173,7 @@ def exception_type(
elif original_exception.status_code == 504: # gateway timeout error
exception_mapping_worked = True
raise Timeout(
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
model=model,
llm_provider="azure",
)
@ -9159,7 +9181,7 @@ def exception_type(
exception_mapping_worked = True
raise APIError(
status_code=original_exception.status_code,
message=f"AzureException - {original_exception.message}",
message=f"AzureException - {original_exception.message} {extra_information}",
llm_provider="azure",
model=model,
request=httpx.Request(
@ -9169,7 +9191,7 @@ def exception_type(
else:
# if no status code then it is an APIConnectionError: https://github.com/openai/openai-python#handling-errors
raise APIConnectionError(
message=f"{exception_provider} - {message}",
message=f"{exception_provider} - {message} {extra_information}",
llm_provider="azure",
model=model,
request=httpx.Request(method="POST", url="https://openai.com/"),