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

docs(exception_mapping.md): update exception mapping docs with 'should_retry'

Browse source
This commit is contained in:
Krrish Dholakia 2024-05-03 16:51:12 -07:00
parent 2084cfd959
commit f7eee60943
2 changed files with 60 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

67
docs/my-website/docs/exception_mapping.md
View file

@ -13,7 +13,7 @@ LiteLLM maps exceptions across all providers to their OpenAI counterparts.
| >=500 | InternalServerError | | >=500 | InternalServerError |
| N/A | ContextWindowExceededError| | N/A | ContextWindowExceededError|
| 400 | ContentPolicyViolationError| | 400 | ContentPolicyViolationError|
| N/A | APIConnectionError | | 500 | APIConnectionError |
Base case we return APIConnectionError Base case we return APIConnectionError
@ -74,6 +74,28 @@ except Exception as e:
``` ```
## Usage - Should you retry exception?
```
import litellm
import openai
try:
response = litellm.completion(
model="gpt-4",
messages=[
{
"role": "user",
"content": "hello, write a 20 pageg essay"
}
],
timeout=0.01, # this will raise a timeout exception
)
except openai.APITimeoutError as e:
should_retry = litellm._should_retry(e.status_code)
print(f"should_retry: {should_retry}")
```
## Details ## Details
To see how it's implemented - [check out the code](https://github.com/BerriAI/litellm/blob/a42c197e5a6de56ea576c73715e6c7c6b19fa249/litellm/utils.py#L1217) To see how it's implemented - [check out the code](https://github.com/BerriAI/litellm/blob/a42c197e5a6de56ea576c73715e6c7c6b19fa249/litellm/utils.py#L1217)
@ -86,21 +108,34 @@ To see how it's implemented - [check out the code](https://github.com/BerriAI/li
Base case - we return the original exception. Base case - we return the original exception.
| | ContextWindowExceededError | AuthenticationError | InvalidRequestError | RateLimitError | ServiceUnavailableError | | custom_llm_provider | Timeout | ContextWindowExceededError | BadRequestError | NotFoundError | ContentPolicyViolationError | AuthenticationError | APIError | RateLimitError | ServiceUnavailableError | PermissionDeniedError | UnprocessableEntityError |
|---------------|----------------------------|---------------------|---------------------|---------------|-------------------------| |----------------------------|---------|----------------------------|------------------|---------------|-----------------------------|---------------------|----------|----------------|-------------------------|-----------------------|-------------------------|
| Anthropic | ✅ | ✅ | ✅ | ✅ | | | openai | ✓ | ✓ | ✓ | | ✓ | ✓ | | | | | |
| OpenAI | ✅ | ✅ |✅ |✅ |✅| | text-completion-openai | ✓ | ✓ | ✓ | | ✓ | ✓ | | | | | |
| Azure OpenAI | ✅ | ✅ |✅ |✅ |✅| | custom_openai | ✓ | ✓ | ✓ | | ✓ | ✓ | | | | | |
| Replicate | ✅ | ✅ | ✅ | ✅ | ✅ | | openai_compatible_providers| ✓ | ✓ | ✓ | | ✓ | ✓ | | | | | |
| Cohere | ✅ | ✅ | ✅ | ✅ | ✅ | | anthropic | ✓ | ✓ | ✓ | ✓ | | ✓ | | | ✓ | ✓ | |
| Huggingface | ✅ | ✅ | ✅ | ✅ | | | replicate | ✓ | ✓ | ✓ | ✓ | | ✓ | | ✓ | ✓ | | |
| Openrouter | ✅ | ✅ | ✅ | ✅ | | | bedrock | ✓ | ✓ | ✓ | ✓ | | ✓ | | ✓ | ✓ | ✓ | |
| AI21 | ✅ | ✅ | ✅ | ✅ | | | sagemaker | | ✓ | ✓ | | | | | | | | |
| VertexAI | | |✅ | | | | vertex_ai | ✓ | | ✓ | | | | ✓ | | | | ✓ |
| Bedrock | | |✅ | | | | palm | ✓ | ✓ | | | | | ✓ | | | | |
| Sagemaker | | |✅ | | | | gemini | ✓ | ✓ | | | | | ✓ | | | | |
| TogetherAI | ✅ | ✅ | ✅ | ✅ | | | cloudflare | | | ✓ | | | ✓ | | | | | |
| AlephAlpha | ✅ | ✅ | ✅ | ✅ | ✅ | | cohere | | ✓ | ✓ | | | ✓ | | | ✓ | | |
| cohere_chat | | ✓ | ✓ | | | ✓ | | | ✓ | | |
| huggingface | ✓ | ✓ | ✓ | | | ✓ | | ✓ | ✓ | | |
| ai21 | ✓ | ✓ | ✓ | ✓ | | ✓ | | ✓ | | | |
| nlp_cloud | ✓ | ✓ | ✓ | | | ✓ | ✓ | ✓ | ✓ | | |
| together_ai | ✓ | ✓ | ✓ | | | ✓ | | | | | |
| aleph_alpha | | | ✓ | | | ✓ | | | | | |
| ollama | ✓ | | ✓ | | | | | | ✓ | | |
| ollama_chat | ✓ | | ✓ | | | | | | ✓ | | |
| vllm | | | | | | ✓ | ✓ | | | | |
| azure | ✓ | ✓ | ✓ | ✓ | ✓ | ✓ | | | ✓ | | |
- "✓" indicates that the specified `custom_llm_provider` can raise the corresponding exception.
- Empty cells indicate the lack of association or that the provider does not raise that particular exception type as indicated by the function.
> For a deeper understanding of these exceptions, you can check out [this](https://github.com/BerriAI/litellm/blob/d7e58d13bf9ba9edbab2ab2f096f3de7547f35fa/litellm/utils.py#L1544) implementation for additional insights. > For a deeper understanding of these exceptions, you can check out [this](https://github.com/BerriAI/litellm/blob/d7e58d13bf9ba9edbab2ab2f096f3de7547f35fa/litellm/utils.py#L1544) implementation for additional insights.

15
litellm/utils.py
View file

@ -8480,7 +8480,7 @@ def exception_type(
# 503 Getting metadata from plugin failed with error: Reauthentication is needed. Please run `gcloud auth application-default login` to reauthenticate. # 503 Getting metadata from plugin failed with error: Reauthentication is needed. Please run `gcloud auth application-default login` to reauthenticate.
exception_mapping_worked = True exception_mapping_worked = True
raise BadRequestError( raise BadRequestError(
message=f"PalmException - Invalid api key", message=f"GeminiException - Invalid api key",
model=model, model=model,
llm_provider="palm", llm_provider="palm",
response=original_exception.response, response=original_exception.response,
@ -8491,23 +8491,26 @@ def exception_type(
): ):
exception_mapping_worked = True exception_mapping_worked = True
raise Timeout( raise Timeout(
message=f"PalmException - {original_exception.message}", message=f"GeminiException - {original_exception.message}",
model=model, model=model,
llm_provider="palm", llm_provider="palm",
) )
if "400 Request payload size exceeds" in error_str: if "400 Request payload size exceeds" in error_str:
exception_mapping_worked = True exception_mapping_worked = True
raise ContextWindowExceededError( raise ContextWindowExceededError(
message=f"PalmException - {error_str}", message=f"GeminiException - {error_str}",
model=model, model=model,
llm_provider="palm", llm_provider="palm",
response=original_exception.response, response=original_exception.response,
) )
if "500 An internal error has occurred." in error_str: if (
"500 An internal error has occurred." in error_str
or "list index out of range" in error_str
):
exception_mapping_worked = True exception_mapping_worked = True
raise APIError( raise APIError(
status_code=getattr(original_exception, "status_code", 500), status_code=getattr(original_exception, "status_code", 500),
message=f"PalmException - {original_exception.message}", message=f"GeminiException - {original_exception.message}",
llm_provider="palm", llm_provider="palm",
model=model, model=model,
request=original_exception.request, request=original_exception.request,
@ -8516,7 +8519,7 @@ def exception_type(
if original_exception.status_code == 400: if original_exception.status_code == 400:
exception_mapping_worked = True exception_mapping_worked = True
raise BadRequestError( raise BadRequestError(
message=f"PalmException - {error_str}", message=f"GeminiException - {error_str}",
model=model, model=model,
llm_provider="palm", llm_provider="palm",
response=original_exception.response, response=original_exception.response,