scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-08-28 05:25:59 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 36

chore: migrate Python formatter from black to ruff format (#31317)

Browse source
This commit is contained in:
Mateo Wang 2026-06-25 11:27:43 -07:00 • committed by GitHub
parent 6db55e0aa5
commit 17bfd415ae
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
215 changed files with 1877 additions and 1241 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

4
.github/workflows/test-linting.yml vendored
View file

@ -50,10 +50,10 @@ jobs:
run: |
uv sync --frozen
- name: Check Black formatting
- name: Check ruff format
run: |
cd litellm
uv run --no-sync black --check --exclude '/enterprise/' .
uv run --no-sync ruff format --check --line-length 88 --exclude '/enterprise/' .
cd ..
- name: Debug - Check file state

16
Makefile
View file

@ -20,13 +20,13 @@ help:
@echo " make install-test-deps - Install the full local test environment"
@echo " make install-helm-unittest - Install helm unittest plugin"
@echo " make install-hooks - Install git hooks (Conventional Commits + Branches)"
@echo " make format - Apply Black code formatting"
@echo " make format-check - Check Black code formatting (matches CI)"
@echo " make lint - Run all linting (Ruff, basedpyright, Black check, circular imports, import safety)"
@echo " make format - Apply ruff format code formatting"
@echo " make format-check - Check ruff format code formatting (matches CI)"
@echo " make lint - Run all linting (Ruff, basedpyright, format check, circular imports, import safety)"
@echo " make lint-ruff - Run Ruff linting only"
@echo " make lint-basedpyright - Run basedpyright strict, gated by per-rule error counts"
@echo " make lint-basedpyright-budget-update - Re-capture the basedpyright per-rule budget (ratchet)"
@echo " make lint-black - Check Black formatting (matches CI)"
@echo " make lint-format - Check ruff format formatting (matches CI)"
@echo " make lint-ruff-budget - Gate the codebase total of each strict ruff rule against its ceiling"
@echo " make lint-gate - Strict ruff gate in CI-parity mode (fetches staging, simulates the merge)"
@echo " make lint-ruff-budget-update - Re-capture per-rule baselines in ruff-strict-budget.json (ratchet)"
@ -82,11 +82,13 @@ install-hooks:
./scripts/install_git_hooks.sh
# Formatting
# 88-column wrap matches the Black width the whole repo is formatted to; ruff.toml's
# global line-length is 120 (for E501/isort), so 88 is forced here.
format: install-dev
cd litellm && $(UV_RUN) black . && cd ..
cd litellm && $(UV_RUN) ruff format --line-length 88 --exclude '/enterprise/' . && cd ..
format-check: install-dev
cd litellm && $(UV_RUN) black --check . && cd ..
cd litellm && $(UV_RUN) ruff format --check --line-length 88 --exclude '/enterprise/' . && cd ..
# Linting targets
lint-ruff: install-dev
@ -131,7 +133,7 @@ lint-basedpyright: install-dev
lint-basedpyright-budget-update: install-dev
($(UV_RUN) basedpyright --outputjson || true) | $(UV_RUN) python scripts/type_check_gate.py --update
lint-black: format-check
lint-format: format-check
lint-ruff-budget: install-dev
$(UV_RUN) python scripts/ruff_strict_gate.py

42
litellm/__init__.py
View file

@ -390,12 +390,8 @@ require_managed_files: bool = (
enable_caching_on_provider_specific_optional_params: bool = (
False # feature-flag for caching on optional params - e.g. 'top_k'
)
caching: bool = (
False # Not used anymore, will be removed in next MAJOR release - https://github.com/BerriAI/litellm/discussions/648
)
caching_with_models: bool = (
False # # Not used anymore, will be removed in next MAJOR release - https://github.com/BerriAI/litellm/discussions/648
)
caching: bool = False # Not used anymore, will be removed in next MAJOR release - https://github.com/BerriAI/litellm/discussions/648
caching_with_models: bool = False # # Not used anymore, will be removed in next MAJOR release - https://github.com/BerriAI/litellm/discussions/648
cache: Optional["Cache"] = (
None # cache object <- use this - https://docs.litellm.ai/docs/caching
)
@ -416,9 +412,7 @@ forward_traceparent_to_llm_provider: bool = False
_current_cost = 0.0 # private variable, used if max budget is set
error_logs: Dict = {}
add_function_to_prompt: bool = (
False # if function calling not supported by api, append function call details to system prompt
)
add_function_to_prompt: bool = False # if function calling not supported by api, append function call details to system prompt
client_session: Optional[httpx.Client] = None
aclient_session: Optional[httpx.AsyncClient] = None
model_fallbacks: Optional[List] = None # Deprecated for 'litellm.fallbacks'
@ -485,9 +479,7 @@ prometheus_end_user_metrics_cleanup_interval_seconds: Optional[float] = 60.0
disable_add_prefix_to_prompt: bool = (
False # used by anthropic, to disable adding prefix to prompt
)
disable_copilot_system_to_assistant: bool = (
False # If false (default), converts all 'system' role messages to 'assistant' for GitHub Copilot compatibility. Set to true to disable this behavior.
)
disable_copilot_system_to_assistant: bool = False # If false (default), converts all 'system' role messages to 'assistant' for GitHub Copilot compatibility. Set to true to disable this behavior.
public_mcp_servers: Optional[List[str]] = None
public_mcp_hub_strict_whitelist: bool = True
public_model_groups: Optional[List[str]] = None
@ -507,17 +499,13 @@ if TYPE_CHECKING:
######## Networking Settings ########
use_aiohttp_transport: bool = (
True # Older variable, aiohttp is now the default. use disable_aiohttp_transport instead.
)
use_aiohttp_transport: bool = True # Older variable, aiohttp is now the default. use disable_aiohttp_transport instead.
aiohttp_trust_env: bool = False # set to true to use HTTP_ Proxy settings
disable_aiohttp_transport: bool = False # Set this to true to use httpx instead
disable_aiohttp_trust_env: bool = (
False # When False, aiohttp will respect HTTP(S)_PROXY env vars
)
force_ipv4: bool = (
False # when True, litellm will force ipv4 for all LLM requests. Some users have seen httpx ConnectionError when using ipv6.
)
force_ipv4: bool = False # when True, litellm will force ipv4 for all LLM requests. Some users have seen httpx ConnectionError when using ipv6.
network_mock: bool = False # When True, use mock transport — no real network calls
####### STOP SEQUENCE LIMIT #######
@ -551,12 +539,12 @@ output_parse_pii: bool = False
from litellm.litellm_core_utils.get_model_cost_map import get_model_cost_map
model_cost = get_model_cost_map(url=model_cost_map_url)
cost_discount_config: Dict[str, float] = (
{}
) # Provider-specific cost discounts {"vertex_ai": 0.05} = 5% discount
cost_margin_config: Dict[str, Union[float, Dict[str, float]]] = (
{}
) # Provider-specific or global cost margins. Examples:
cost_discount_config: Dict[
str, float
] = {} # Provider-specific cost discounts {"vertex_ai": 0.05} = 5% discount
cost_margin_config: Dict[
str, Union[float, Dict[str, float]]
] = {} # Provider-specific or global cost margins. Examples:
# Percentage: {"openai": 0.10} = 10% margin
# Fixed: {"openai": {"fixed_amount": 0.001}} = $0.001 per request
# Global: {"global": 0.05} = 5% global margin on all providers
@ -1457,9 +1445,9 @@ from . import rag
from .types.llms.custom_llm import CustomLLMItem
custom_provider_map: List[CustomLLMItem] = []
_custom_providers: List[str] = (
[]
) # internal helper util, used to track names of custom providers
_custom_providers: List[
str
] = [] # internal helper util, used to track names of custom providers
disable_hf_tokenizer_download: Optional[bool] = (
None # disable huggingface tokenizer download. Defaults to openai clk100
)

11
litellm/_redis.py
View file

@ -327,7 +327,9 @@ def _get_redis_client_logic(**env_overrides):
**env_overrides,
}
_startup_nodes: Optional[Union[str, list]] = redis_kwargs.get("startup_nodes", None) or get_secret( # type: ignore
_startup_nodes: Optional[Union[str, list]] = redis_kwargs.get(
"startup_nodes", None
) or get_secret( # type: ignore
"REDIS_CLUSTER_NODES"
)
@ -338,7 +340,9 @@ def _get_redis_client_logic(**env_overrides):
elif _startup_nodes is None:
redis_kwargs.pop("startup_nodes", None)
_sentinel_nodes: Optional[Union[str, list]] = redis_kwargs.get("sentinel_nodes", None) or get_secret( # type: ignore
_sentinel_nodes: Optional[Union[str, list]] = redis_kwargs.get(
"sentinel_nodes", None
) or get_secret( # type: ignore
"REDIS_SENTINEL_NODES"
)
@ -609,7 +613,8 @@ def get_redis_async_client(
# Create async RedisCluster with IAM token as password if available
cluster_client = async_redis.RedisCluster(
startup_nodes=new_startup_nodes, **cluster_kwargs # type: ignore
startup_nodes=new_startup_nodes,
**cluster_kwargs, # type: ignore
)
return cluster_client

32
litellm/assistants/main.py
View file

@ -184,7 +184,9 @@ def get_assistants(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
@ -198,7 +200,9 @@ def get_assistants(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
@ -394,7 +398,9 @@ def create_assistants(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
if response is None:
@ -761,7 +767,9 @@ def create_thread(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response # type: ignore
@ -916,7 +924,9 @@ def get_thread(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response # type: ignore
@ -1103,7 +1113,9 @@ def add_message(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
@ -1263,7 +1275,9 @@ def get_messages(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
@ -1478,7 +1492,9 @@ def run_thread(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response # type: ignore

4
litellm/assistants/utils.py
View file

@ -71,9 +71,7 @@ def get_optional_params_add_message(
if custom_llm_provider == "openai":
optional_params = non_default_params
elif custom_llm_provider == "azure":
supported_params = (
litellm.AzureOpenAIAssistantsAPIConfig().get_supported_openai_create_message_params()
)
supported_params = litellm.AzureOpenAIAssistantsAPIConfig().get_supported_openai_create_message_params()
_check_valid_arg(supported_params=supported_params)
optional_params = litellm.AzureOpenAIAssistantsAPIConfig().map_openai_params_create_message_params(
non_default_params=non_default_params, optional_params=optional_params

22
litellm/batches/main.py
View file

@ -359,7 +359,9 @@ def create_batch(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_batch", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_batch", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -553,7 +555,9 @@ def _handle_retrieve_batch_providers_without_provider_config(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="retrieve_batch", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="retrieve_batch", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -819,7 +823,11 @@ def list_batches(
max_retries=optional_params.max_retries,
)
elif custom_llm_provider == "azure":
api_base = optional_params.api_base or litellm.api_base or get_secret_str("AZURE_API_BASE") # type: ignore
api_base = (
optional_params.api_base
or litellm.api_base
or get_secret_str("AZURE_API_BASE")
) # type: ignore
api_version = (
optional_params.api_version
or litellm.api_version
@ -887,7 +895,9 @@ def list_batches(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -1097,7 +1107,9 @@ def cancel_batch(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="cancel_batch", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="cancel_batch", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response

4
litellm/budget_manager.py
View file

@ -67,9 +67,7 @@ class BudgetManager:
)
response = response.json()
if response["status"] == "error":
self.user_dict = (
{}
) # assume this means the user dict hasn't been stored yet
self.user_dict = {} # assume this means the user dict hasn't been stored yet
else:
self.user_dict = response["data"]

13
litellm/caching/caching_handler.py
View file

@ -79,9 +79,7 @@ class CachingHandlerResponse(BaseModel):
cached_result: Optional[Any] = None
final_embedding_cached_response: Optional[EmbeddingResponse] = None
embedding_all_elements_cache_hit: bool = (
False # this is set to True when all elements in the list have a cache hit in the embedding cache, if true return the final_embedding_cached_response no need to make an API call
)
embedding_all_elements_cache_hit: bool = False # this is set to True when all elements in the list have a cache hit in the embedding cache, if true return the final_embedding_cached_response no need to make an API call
in_memory_cache_obj = InMemoryCache()
@ -165,10 +163,11 @@ class LLMCachingHandler:
"""
# Check if caching should be performed BEFORE doing expensive operations
if (
(kwargs.get("caching", None) is None and litellm.cache is not None)
or kwargs.get("caching", False) is True
) and (
kwargs.get("cache", {}).get("no-cache", False) is not True
(
(kwargs.get("caching", None) is None and litellm.cache is not None)
or kwargs.get("caching", False) is True
)
and (kwargs.get("cache", {}).get("no-cache", False) is not True)
): # allow users to control returning cached responses from the completion function
args = args or ()
final_embedding_cached_response: Optional[EmbeddingResponse] = None

3
litellm/caching/redis_cluster_cache.py
View file

@ -79,7 +79,8 @@ class RedisClusterCache(RedisCache):
# Create a fresh Redis Cluster client with current settings
redis_client = redis_async.RedisCluster(
startup_nodes=new_startup_nodes, **cluster_kwargs # type: ignore
startup_nodes=new_startup_nodes,
**cluster_kwargs, # type: ignore
)
# Test the connection

4
litellm/completion_extras/litellm_responses_transformation/handler.py
View file

@ -151,7 +151,9 @@ class ResponsesToCompletionBridgeHandler:
custom_llm_provider=custom_llm_provider,
)
def completion(self, *args, **kwargs) -> Union[
def completion(
self, *args, **kwargs
) -> Union[
Coroutine[Any, Any, Union["ModelResponse", "CustomStreamWrapper"]],
"ModelResponse",
"CustomStreamWrapper",

22
litellm/completion_extras/litellm_responses_transformation/transformation.py
View file

@ -205,9 +205,9 @@ class LiteLLMResponsesTransformationHandler(CompletionTransformationBridge):
if provider_specific_fields:
tool_call_dict["provider_specific_fields"] = provider_specific_fields
# Also add to function's provider_specific_fields for consistency
tool_call_dict["function"][
"provider_specific_fields"
] = provider_specific_fields
tool_call_dict["function"]["provider_specific_fields"] = (
provider_specific_fields
)
msg = Message(
content=None,
@ -301,7 +301,9 @@ class LiteLLMResponsesTransformationHandler(CompletionTransformationBridge):
{
"type": "message",
"role": role,
"content": self._convert_content_to_responses_format(content, cast(str, role)), # type: ignore[arg-type]
"content": self._convert_content_to_responses_format(
content, cast(str, role)
), # type: ignore[arg-type]
}
)
@ -1021,7 +1023,11 @@ class LiteLLMResponsesTransformationHandler(CompletionTransformationBridge):
# If string is passed, map with optional summary based on flag/env var
if reasoning_effort == "none":
return Reasoning(effort="none", summary="detailed") if auto_summary_enabled else Reasoning(effort="none") # type: ignore
return (
Reasoning(effort="none", summary="detailed")
if auto_summary_enabled
else Reasoning(effort="none")
) # type: ignore
elif reasoning_effort == "high":
return (
Reasoning(effort="high", summary="detailed")
@ -1029,7 +1035,11 @@ class LiteLLMResponsesTransformationHandler(CompletionTransformationBridge):
else Reasoning(effort="high")
)
elif reasoning_effort == "xhigh":
return Reasoning(effort="xhigh", summary="detailed") if auto_summary_enabled else Reasoning(effort="xhigh") # type: ignore[typeddict-item]
return (
Reasoning(effort="xhigh", summary="detailed")
if auto_summary_enabled
else Reasoning(effort="xhigh")
) # type: ignore[typeddict-item]
elif reasoning_effort == "medium":
return (
Reasoning(effort="medium", summary="detailed")

51
litellm/constants.py
View file

@ -296,7 +296,8 @@ DEFAULT_SSL_CIPHERS = os.getenv(
"ECDHE-ECDSA-AES256-GCM-SHA384:"
"ECDHE-ECDSA-AES128-GCM-SHA256:"
# Priority 3: Additional modern ciphers (good balance)
"ECDHE-RSA-CHACHA20-POLY1305:" "ECDHE-ECDSA-CHACHA20-POLY1305:"
"ECDHE-RSA-CHACHA20-POLY1305:"
"ECDHE-ECDSA-CHACHA20-POLY1305:"
# Priority 4: Widely compatible fallbacks (slower but universally supported)
"ECDHE-RSA-AES256-SHA384:" # Common fallback
"ECDHE-RSA-AES128-SHA256:" # Very widely supported
@ -885,31 +886,29 @@ openai_compatible_providers: List = [
"pinstripes", # Pinstripes - JSON-configured provider
"darkbloom",
]
openai_text_completion_compatible_providers: List = (
[ # providers that support `/v1/completions`
"together_ai",
"fireworks_ai",
"hosted_vllm",
"meta_llama",
"llamafile",
"featherless_ai",
"nebius",
"dashscope",
"modelscope",
"moonshot",
"publicai",
"synthetic",
"tensormesh",
"apertis",
"nano-gpt",
"poe",
"chutes",
"v0",
"lambda_ai",
"hyperbolic",
"wandb",
]
)
openai_text_completion_compatible_providers: List = [ # providers that support `/v1/completions`
"together_ai",
"fireworks_ai",
"hosted_vllm",
"meta_llama",
"llamafile",
"featherless_ai",
"nebius",
"dashscope",
"modelscope",
"moonshot",
"publicai",
"synthetic",
"tensormesh",
"apertis",
"nano-gpt",
"poe",
"chutes",
"v0",
"lambda_ai",
"hyperbolic",
"wandb",
]
_openai_like_providers: List = [
"predibase",
"databricks",

8
litellm/cost_calculator.py
View file

@ -1020,7 +1020,7 @@ def _apply_cost_discount(
if verbose_logger.isEnabledFor(logging.DEBUG):
verbose_logger.debug(
f"Applied {discount_percent*100}% discount to {custom_llm_provider}: "
f"Applied {discount_percent * 100}% discount to {custom_llm_provider}: "
f"${original_cost:.6f} -> ${final_cost:.6f} (saved ${discount_amount:.6f})"
)
@ -1088,7 +1088,7 @@ def _apply_cost_margin(
verbose_logger.debug(
f"Applied margin to {custom_llm_provider or 'global'}: "
f"${original_cost:.6f} -> ${final_cost:.6f} "
f"(margin: {margin_percent*100 if margin_percent > 0 else 0}% + ${margin_fixed_amount:.6f} = ${margin_total_amount:.6f})"
f"(margin: {margin_percent * 100 if margin_percent > 0 else 0}% + ${margin_fixed_amount:.6f} = ${margin_total_amount:.6f})"
)
return final_cost, margin_percent, margin_fixed_amount, margin_total_amount
@ -1621,7 +1621,9 @@ def completion_cost(
model in litellm.replicate_models or "replicate" in model
) and model not in litellm.model_cost:
# for unmapped replicate model, default to replicate's time tracking logic
return get_replicate_completion_pricing(completion_response, total_time) # type: ignore
return get_replicate_completion_pricing(
completion_response, total_time
) # type: ignore
if model is None:
raise ValueError(

26
litellm/files/main.py
View file

@ -264,7 +264,9 @@ def create_file(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_file", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_file", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -435,7 +437,10 @@ def file_retrieve(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread",
url="https://github.com/BerriAI/litellm",
), # type: ignore
),
)
@ -618,7 +623,10 @@ def file_delete(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread",
url="https://github.com/BerriAI/litellm",
), # type: ignore
),
)
return cast(FileDeleted, response)
@ -786,7 +794,9 @@ def file_list(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="file_list", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="file_list", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -1037,7 +1047,9 @@ def file_content(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -1118,7 +1130,9 @@ def file_content_streaming(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)

41
litellm/fine_tuning/main.py
View file

@ -245,7 +245,11 @@ def create_fine_tuning_job(
)
# Azure OpenAI
elif custom_llm_provider == "azure":
api_base = optional_params.api_base or litellm.api_base or get_secret_str("AZURE_API_BASE") # type: ignore
api_base = (
optional_params.api_base
or litellm.api_base
or get_secret_str("AZURE_API_BASE")
) # type: ignore
api_version = (
optional_params.api_version
@ -340,7 +344,9 @@ def create_fine_tuning_job(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -458,7 +464,11 @@ def cancel_fine_tuning_job(
)
# Azure OpenAI
elif custom_llm_provider == "azure":
api_base = optional_params.api_base or litellm.api_base or get_secret("AZURE_API_BASE") # type: ignore
api_base = (
optional_params.api_base
or litellm.api_base
or get_secret("AZURE_API_BASE")
) # type: ignore
api_version = (
optional_params.api_version
@ -500,7 +510,9 @@ def cancel_fine_tuning_job(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -621,7 +633,11 @@ def list_fine_tuning_jobs(
)
# Azure OpenAI
elif custom_llm_provider == "azure":
api_base = optional_params.api_base or litellm.api_base or get_secret_str("AZURE_API_BASE") # type: ignore
api_base = (
optional_params.api_base
or litellm.api_base
or get_secret_str("AZURE_API_BASE")
) # type: ignore
api_version = (
optional_params.api_version
@ -664,7 +680,9 @@ def list_fine_tuning_jobs(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="create_thread", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="create_thread", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
return response
@ -776,7 +794,11 @@ def retrieve_fine_tuning_job(
)
# Azure OpenAI
elif custom_llm_provider == "azure":
api_base = optional_params.api_base or litellm.api_base or get_secret_str("AZURE_API_BASE") # type: ignore
api_base = (
optional_params.api_base
or litellm.api_base
or get_secret_str("AZURE_API_BASE")
) # type: ignore
api_version = (
optional_params.api_version
@ -818,7 +840,10 @@ def retrieve_fine_tuning_job(
response=httpx.Response(
status_code=400,
content="Unsupported provider",
request=httpx.Request(method="retrieve_fine_tuning_job", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="retrieve_fine_tuning_job",
url="https://github.com/BerriAI/litellm",
), # type: ignore
),
)
return response

6
litellm/google_genai/adapters/transformation.py
View file

@ -726,9 +726,9 @@ class GoogleGenAIAdapter:
wrapper.accumulated_tool_calls[tool_call_index]["name"] = function_name
if args_chunk:
wrapper.accumulated_tool_calls[tool_call_index][
"arguments"
] += args_chunk
wrapper.accumulated_tool_calls[tool_call_index]["arguments"] += (
args_chunk
)
# Attempt to parse and emit a complete tool call
accumulated_data = wrapper.accumulated_tool_calls[tool_call_index]

24
litellm/images/main.py
View file

@ -880,20 +880,20 @@ def image_edit(
local_vars.update(kwargs)
# Get ImageEditOptionalRequestParams with only valid parameters
image_edit_optional_params: (
ImageEditOptionalRequestParams
) = _get_ImageEditRequestUtils().get_requested_image_edit_optional_param(
local_vars
image_edit_optional_params: ImageEditOptionalRequestParams = (
_get_ImageEditRequestUtils().get_requested_image_edit_optional_param(
local_vars
)
)
# Get optional parameters for the responses API
image_edit_request_params: (
Dict
) = _get_ImageEditRequestUtils().get_optional_params_image_edit(
model=model,
image_edit_provider_config=image_edit_provider_config,
image_edit_optional_params=image_edit_optional_params,
drop_params=kwargs.get("drop_params"),
additional_drop_params=kwargs.get("additional_drop_params"),
image_edit_request_params: Dict = (
_get_ImageEditRequestUtils().get_optional_params_image_edit(
model=model,
image_edit_provider_config=image_edit_provider_config,
image_edit_optional_params=image_edit_optional_params,
drop_params=kwargs.get("drop_params"),
additional_drop_params=kwargs.get("additional_drop_params"),
)
)
# Pre Call logging

8
litellm/integrations/SlackAlerting/hanging_request_check.py
View file

@ -104,10 +104,10 @@ class AlertingHangingRequestCheck:
)
for request_id in hanging_requests:
hanging_request_data: Optional[HangingRequestData] = (
await self.hanging_request_cache.async_get_cache(
key=request_id,
)
hanging_request_data: Optional[
HangingRequestData
] = await self.hanging_request_cache.async_get_cache(
key=request_id,
)
if hanging_request_data is None:

22
litellm/integrations/SlackAlerting/slack_alerting.py
View file

@ -245,7 +245,7 @@ class SlackAlerting(CustomBatchLogger):
return
for api_base, latency in _deployment_latency_map.items():
_message_to_send += f"\n{api_base}: {round(latency,2)}s"
_message_to_send += f"\n{api_base}: {round(latency, 2)}s"
_message_to_send = "```" + _message_to_send + "```"
return _message_to_send
@ -272,7 +272,7 @@ class SlackAlerting(CustomBatchLogger):
if litellm.turn_off_message_logging or litellm.redact_messages_in_exceptions:
messages = "Message not logged. litellm.redact_messages_in_exceptions=True"
request_info = f"\nRequest Model: `{model}`\nAPI Base: `{api_base}`\nMessages: `{messages}`"
slow_message = f"`Responses are slow - {round(time_difference_float,2)}s response time > Alerting threshold: {self.alerting_threshold}s`"
slow_message = f"`Responses are slow - {round(time_difference_float, 2)}s response time > Alerting threshold: {self.alerting_threshold}s`"
alerting_metadata: dict = {}
if time_difference_float > self.alerting_threshold:
# add deployment latencies to alert
@ -460,7 +460,7 @@ class SlackAlerting(CustomBatchLogger):
if api_base is None:
api_base = ""
value = replaced_failed_values[top_5_failed[i]]
message += f"\t{i+1}. Deployment: `{deployment_name}`, Failed Requests: `{value}`, API Base: `{api_base}`\n"
message += f"\t{i + 1}. Deployment: `{deployment_name}`, Failed Requests: `{value}`, API Base: `{api_base}`\n"
message += "\n\n*😅 Top Slowest Deployments:*\n\n"
if not top_5_slowest:
@ -479,7 +479,7 @@ class SlackAlerting(CustomBatchLogger):
),
)
value = round(replaced_slowest_values[top_5_slowest[i]], 3)
message += f"\t{i+1}. Deployment: `{deployment_name}`, Latency per output token: `{value}s/token`, API Base: `{api_base}`\n\n"
message += f"\t{i + 1}. Deployment: `{deployment_name}`, Latency per output token: `{value}s/token`, API Base: `{api_base}`\n\n"
# cache cleanup -> reset values to 0
latency_cache_keys = [(key, 0) for key in latency_keys]
@ -595,9 +595,7 @@ class SlackAlerting(CustomBatchLogger):
"projected_limit_exceeded",
"soft_budget_crossed",
]
] = (
"projected_limit_exceeded" if type == "projected_limit_exceeded" else None
)
] = "projected_limit_exceeded" if type == "projected_limit_exceeded" else None
webhook_event: Optional[WebhookEvent] = None
@ -854,9 +852,9 @@ class SlackAlerting(CustomBatchLogger):
### UNIQUE CACHE KEY ###
cache_key = provider + region_name
outage_value: Optional[ProviderRegionOutageModel] = (
await self.internal_usage_cache.async_get_cache(key=cache_key)
)
outage_value: Optional[
ProviderRegionOutageModel
] = await self.internal_usage_cache.async_get_cache(key=cache_key)
# Convert deployment_ids back to set if it was stored as a list
if outage_value is not None:
@ -981,7 +979,9 @@ class SlackAlerting(CustomBatchLogger):
max_alerts_size = 10
"""
try:
outage_value: Optional[OutageModel] = await self.internal_usage_cache.async_get_cache(key=deployment_id) # type: ignore
outage_value: Optional[
OutageModel
] = await self.internal_usage_cache.async_get_cache(key=deployment_id) # type: ignore
if (
getattr(exception, "status_code", None) is None
or (

6
litellm/integrations/braintrust_logging.py
View file

@ -52,9 +52,9 @@ class BraintrustLogger(CustomLogger):
"Authorization": "Bearer " + self.api_key,
"Content-Type": "application/json",
}
self._project_id_cache: Dict[str, str] = (
{}
) # Cache mapping project names to IDs
self._project_id_cache: Dict[
str, str
] = {} # Cache mapping project names to IDs
self.global_braintrust_http_handler = get_async_httpx_client(
llm_provider=httpxSpecialProvider.LoggingCallback
)

2
litellm/integrations/braintrust_mock_client.py
View file

@ -157,7 +157,7 @@ def create_mock_braintrust_client():
create_mock_braintrust_factory_client()
verbose_logger.debug(
f"[BRAINTRUST MOCK] Mock latency set to {_MOCK_LATENCY_SECONDS*1000:.0f}ms"
f"[BRAINTRUST MOCK] Mock latency set to {_MOCK_LATENCY_SECONDS * 1000:.0f}ms"
)
verbose_logger.debug(
"[BRAINTRUST MOCK] Braintrust mock client initialization complete"

4
litellm/integrations/custom_logger.py
View file

@ -981,7 +981,9 @@ class CustomLogger: # https://docs.litellm.ai/docs/observability/custom_callbac
verbose_logger.debug(
f"Incrementing callback failure metric for {callback_name}"
)
callback_obj.increment_callback_logging_failure(callback_name=callback_name) # type: ignore
callback_obj.increment_callback_logging_failure(
callback_name=callback_name
) # type: ignore
return
verbose_logger.debug(

4
litellm/integrations/datadog/datadog_metrics.py
View file

@ -256,7 +256,9 @@ class DatadogMetricsLogger(CustomBatchLogger):
headers["Content-Encoding"] = "gzip"
response = await self.async_client.post(
self.upload_url, content=compressed_data, headers=headers # type: ignore
self.upload_url,
content=compressed_data,
headers=headers, # type: ignore
)
response.raise_for_status()

2
litellm/integrations/focus/destinations/vantage_destination.py
View file

@ -157,7 +157,7 @@ class FocusVantageDestination(FocusDestination):
async def _upload_csv(
self, client: AsyncHTTPHandler, csv_bytes: bytes, filename: str
) -> None:
url = f"{self.base_url}/v2/integrations/" f"{self.integration_token}/costs.csv"
url = f"{self.base_url}/v2/integrations/{self.integration_token}/costs.csv"
headers = {
"Authorization": f"Bearer {self.api_key}",
}

7
litellm/integrations/gcs_bucket/gcs_bucket_mock_client.py
View file

@ -154,7 +154,10 @@ def create_mock_gcs_client():
This function is idempotent - it only initializes mocks once, even if called multiple times.
"""
global _original_async_handler_get, _original_async_handler_delete, _mocks_initialized
global \
_original_async_handler_get, \
_original_async_handler_delete, \
_mocks_initialized
# Use factory for POST handler
_create_mock_gcs_post()
@ -179,7 +182,7 @@ def create_mock_gcs_client():
verbose_logger.debug("[GCS MOCK] Patched AsyncHTTPHandler.delete")
verbose_logger.debug(
f"[GCS MOCK] Mock latency set to {_MOCK_LATENCY_SECONDS*1000:.0f}ms"
f"[GCS MOCK] Mock latency set to {_MOCK_LATENCY_SECONDS * 1000:.0f}ms"
)
verbose_logger.debug("[GCS MOCK] GCS mock client initialization complete")

29
litellm/integrations/gitlab/gitlab_prompt_manager.py
View file

@ -372,7 +372,9 @@ class GitLabPromptManager(CustomPromptManagement):
if parsed_messages:
final_messages: List[AllMessageValues] = parsed_messages
else:
final_messages = [{"role": "user", "content": rendered_prompt}] + messages # type: ignore
final_messages = [
{"role": "user", "content": rendered_prompt}
] + messages # type: ignore
if litellm_params is None:
litellm_params = {}
@ -412,24 +414,41 @@ class GitLabPromptManager(CustomPromptManagement):
low = line.lower()
if low.startswith("system:"):
if current_role and current_content:
messages.append({"role": current_role, "content": "\n".join(current_content).strip()}) # type: ignore
messages.append(
{
"role": current_role,
"content": "\n".join(current_content).strip(),
}
) # type: ignore
current_role = "system"
current_content = [line[7:].strip()]
elif low.startswith("user:"):
if current_role and current_content:
messages.append({"role": current_role, "content": "\n".join(current_content).strip()}) # type: ignore
messages.append(
{
"role": current_role,
"content": "\n".join(current_content).strip(),
}
) # type: ignore
current_role = "user"
current_content = [line[5:].strip()]
elif low.startswith("assistant:"):
if current_role and current_content:
messages.append({"role": current_role, "content": "\n".join(current_content).strip()}) # type: ignore
messages.append(
{
"role": current_role,
"content": "\n".join(current_content).strip(),
}
) # type: ignore
current_role = "assistant"
current_content = [line[10:].strip()]
else:
current_content.append(line)
if current_role and current_content:
messages.append({"role": current_role, "content": "\n".join(current_content).strip()}) # type: ignore
messages.append(
{"role": current_role, "content": "\n".join(current_content).strip()}
) # type: ignore
if not messages and prompt_content.strip():
messages = [{"role": "user", "content": prompt_content.strip()}] # type: ignore
return messages

14
litellm/integrations/lago.py
View file

@ -131,9 +131,9 @@ class LagoLogger(CustomLogger):
def log_success_event(self, kwargs, response_obj, start_time, end_time):
_url = os.getenv("LAGO_API_BASE")
assert _url is not None and isinstance(
_url, str
), "LAGO_API_BASE missing or not set correctly. LAGO_API_BASE={}".format(_url)
assert _url is not None and isinstance(_url, str), (
"LAGO_API_BASE missing or not set correctly. LAGO_API_BASE={}".format(_url)
)
if _url.endswith("/"):
_url += "api/v1/events"
else:
@ -165,10 +165,10 @@ class LagoLogger(CustomLogger):
try:
verbose_logger.debug("ENTERS LAGO CALLBACK")
_url = os.getenv("LAGO_API_BASE")
assert _url is not None and isinstance(
_url, str
), "LAGO_API_BASE missing or not set correctly. LAGO_API_BASE={}".format(
_url
assert _url is not None and isinstance(_url, str), (
"LAGO_API_BASE missing or not set correctly. LAGO_API_BASE={}".format(
_url
)
)
if _url.endswith("/"):
_url += "api/v1/events"

6
litellm/integrations/langfuse/langfuse_handler.py
View file

@ -86,9 +86,9 @@ class LangFuseHandler:
if globalLangfuseLogger is not None:
return globalLangfuseLogger
credentials_dict: Dict[str, Any] = (
{}
) # the global langfuse logger uses Environment Variables, there are no dynamic credentials
credentials_dict: Dict[
str, Any
] = {} # the global langfuse logger uses Environment Variables, there are no dynamic credentials
globalLangfuseLogger = in_memory_dynamic_logger_cache.get_cache(
credentials=credentials_dict,
service_name="langfuse",

3
litellm/integrations/langsmith.py
View file

@ -65,8 +65,7 @@ class LangsmithLogger(CustomBatchLogger):
langsmith_tenant_id=langsmith_tenant_id,
)
self.sampling_rate: float = (
langsmith_sampling_rate
or float(os.getenv("LANGSMITH_SAMPLING_RATE")) # type: ignore
langsmith_sampling_rate or float(os.getenv("LANGSMITH_SAMPLING_RATE")) # type: ignore
if os.getenv("LANGSMITH_SAMPLING_RATE") is not None
and os.getenv("LANGSMITH_SAMPLING_RATE").strip().isdigit() # type: ignore
else 1.0

8
litellm/integrations/mock_client_factory.py
View file

@ -232,7 +232,11 @@ def create_mock_client_factory(config: MockClientConfig):
# Create mock client initialization function
def create_mock_client():
"""Initialize the mock client by patching HTTP handlers."""
nonlocal _original_async_handler_post, _original_sync_client_post, _original_http_handler_post, _mocks_initialized
nonlocal \
_original_async_handler_post, \
_original_sync_client_post, \
_original_http_handler_post, \
_mocks_initialized
if _mocks_initialized:
return
@ -261,7 +265,7 @@ def create_mock_client_factory(config: MockClientConfig):
verbose_logger.debug(f"[{config.name} MOCK] Patched HTTPHandler.post")
verbose_logger.debug(
f"[{config.name} MOCK] Mock latency set to {_MOCK_LATENCY_SECONDS*1000:.0f}ms"
f"[{config.name} MOCK] Mock latency set to {_MOCK_LATENCY_SECONDS * 1000:.0f}ms"
)
verbose_logger.debug(
f"[{config.name} MOCK] {config.name} mock client initialization complete"

8
litellm/integrations/opik/opik.py
View file

@ -174,7 +174,9 @@ class OpikLogger(CustomBatchLogger):
) -> None:
try:
response = self.sync_httpx_client.post(
url=url, headers=headers, json=batch # type: ignore
url=url,
headers=headers,
json=batch, # type: ignore
)
response.raise_for_status()
if response.status_code != 204:
@ -264,7 +266,9 @@ class OpikLogger(CustomBatchLogger):
) -> None:
try:
response = await self.async_httpx_client.post(
url=url, headers=headers, json=batch # type: ignore
url=url,
headers=headers,
json=batch, # type: ignore
)
response.raise_for_status()

2
litellm/integrations/otel/mappers/__init__.py
View file

@ -38,7 +38,7 @@ def resolve_mappers(names: Iterable[str]) -> list[AttributeMapper]:
factory = _MAPPER_BY_NAME.get(name)
if factory is None:
raise ValueError(
f"unknown mapper name {name!r}; known: " f"{sorted(_MAPPER_BY_NAME)}"
f"unknown mapper name {name!r}; known: {sorted(_MAPPER_BY_NAME)}"
)
out.append(factory())
return out

9
litellm/integrations/otel/mappers/genai.py
View file

@ -35,7 +35,6 @@ from litellm.integrations.otel.model.spans import db_system
class GenAIMapper:
_LLM_CALL_ATTRS: dict[str, Callable[[LLMCallSpanData], AttrValue | None]] = {
GenAI.OPERATION_NAME: lambda d: d.operation.value,
GenAI.PROVIDER_NAME: lambda d: d.provider or None,
@ -81,9 +80,13 @@ class GenAIMapper:
f"{LiteLLM.COST_PREFIX}original": lambda d: d.cost.original,
f"{LiteLLM.COST_PREFIX}discount_amount": lambda d: d.cost.discount_amount,
f"{LiteLLM.COST_PREFIX}discount_percent": lambda d: d.cost.discount_percent,
f"{LiteLLM.COST_PREFIX}margin_fixed_amount": lambda d: d.cost.margin_fixed_amount,
f"{LiteLLM.COST_PREFIX}margin_fixed_amount": lambda d: (
d.cost.margin_fixed_amount
),
f"{LiteLLM.COST_PREFIX}margin_percent": lambda d: d.cost.margin_percent,
f"{LiteLLM.COST_PREFIX}margin_total_amount": lambda d: d.cost.margin_total_amount,
f"{LiteLLM.COST_PREFIX}margin_total_amount": lambda d: (
d.cost.margin_total_amount
),
LiteLLM.REQUEST_STREAMING: lambda d: d.is_streaming,
}

1
litellm/integrations/otel/mappers/langfuse.py
View file

@ -27,7 +27,6 @@ from litellm.integrations.otel.model.payloads import (
class LangfuseMapper:
_LLM_CALL_ATTRS: dict[str, Callable[[LLMCallSpanData], AttrValue | None]] = {
"langfuse.observation.type": lambda d: "generation",
"langfuse.observation.model.name": lambda d: d.request_model or None,

1
litellm/integrations/otel/mappers/langtrace.py
View file

@ -20,7 +20,6 @@ from litellm.integrations.otel.model.payloads import LLMCallSpanData
class LangtraceMapper:
_LLM_CALL_ATTRS: dict[str, Callable[[LLMCallSpanData], AttrValue | None]] = {
"gen_ai.operation.name": lambda d: "chat",
"langtrace.service.name": lambda d: d.provider or None,

8
litellm/integrations/otel/model/baggage.py
View file

@ -29,13 +29,15 @@ _PROMOTABLE: Final[
] = {
LiteLLM.TEAM_ID: lambda identity, model, team_metadata_keys: identity.team_id,
LiteLLM.TEAM_ALIAS: lambda identity, model, team_metadata_keys: identity.team_alias,
LiteLLM.TEAM_METADATA: lambda identity, model, team_metadata_keys: _filtered_team_metadata_json(
identity.team_metadata, team_metadata_keys
LiteLLM.TEAM_METADATA: lambda identity, model, team_metadata_keys: (
_filtered_team_metadata_json(identity.team_metadata, team_metadata_keys)
),
LiteLLM.KEY_HASH: lambda identity, model, team_metadata_keys: identity.key_hash,
LiteLLM.END_USER: lambda identity, model, team_metadata_keys: identity.end_user,
GenAI.REQUEST_MODEL: lambda identity, model, team_metadata_keys: model,
LiteLLM.PROVIDER_MODEL: lambda identity, model, team_metadata_keys: identity.provider_model,
LiteLLM.PROVIDER_MODEL: lambda identity, model, team_metadata_keys: (
identity.provider_model
),
}
# Keys promoted by default (a subset of ``_PROMOTABLE``). ``END_USER`` is

4
litellm/integrations/prometheus.py
View file

@ -3237,7 +3237,9 @@ class PrometheusLogger(CustomLogger):
)
return
async def fetch_keys(page_size: int, page: int) -> Tuple[
async def fetch_keys(
page_size: int, page: int
) -> Tuple[
List[Union[str, UserAPIKeyAuth, LiteLLM_DeletedVerificationToken]],
Optional[int],
]:

3
litellm/integrations/rubrik.py
View file

@ -550,8 +550,7 @@ class RubrikLogger(CustomGuardrail, CustomBatchLogger):
"response": response_data,
}
verbose_logger.debug(
f"Sending request to tool blocking service: "
f"{self.tool_blocking_endpoint}"
f"Sending request to tool blocking service: {self.tool_blocking_endpoint}"
)
http_response = await self.tool_blocking_client.post(
self.tool_blocking_endpoint,

12
litellm/integrations/vector_store_integrations/vector_store_pre_call_hook.py
View file

@ -88,12 +88,12 @@ class VectorStorePreCallHook(CustomLogger):
pass
# Use database fallback to ensure synchronization across instances
vector_stores_to_run: List[LiteLLM_ManagedVectorStore] = (
await litellm.vector_store_registry.pop_vector_stores_to_run_with_db_fallback(
non_default_params=non_default_params,
tools=tools,
prisma_client=prisma_client,
)
vector_stores_to_run: List[
LiteLLM_ManagedVectorStore
] = await litellm.vector_store_registry.pop_vector_stores_to_run_with_db_fallback(
non_default_params=non_default_params,
tools=tools,
prisma_client=prisma_client,
)
if not vector_stores_to_run:

6
litellm/litellm_core_utils/core_helpers.py
View file

@ -361,9 +361,9 @@ def safe_deep_copy(data):
"litellm_metadata" in data
and "litellm_parent_otel_span" in data["litellm_metadata"]
):
data["litellm_metadata"][
"litellm_parent_otel_span"
] = litellm_parent_otel_span
data["litellm_metadata"]["litellm_parent_otel_span"] = (
litellm_parent_otel_span
)
return new_data

21
litellm/litellm_core_utils/exception_mapping_utils.py
View file

@ -272,8 +272,8 @@ def _map_openai_exception(
else:
message = str(original_exception)
if message is not None and isinstance(
message, str
if (
message is not None and isinstance(message, str)
): # done to prevent user-confusion. Relevant issue - https://github.com/BerriAI/litellm/issues/1414
message = message.replace("OPENAI", custom_llm_provider.upper())
message = message.replace(
@ -726,7 +726,6 @@ def _map_openai_like_exception(
extra_information: str,
) -> None:
if "authorization denied for" in error_str:
# Predibase returns the raw API Key in the response - this block ensures it's not returned in the exception
if (
error_str is not None
@ -1161,7 +1160,9 @@ def _map_vertex_exception(
response=httpx.Response(
status_code=500,
content=str(original_exception),
request=httpx.Request(method="completion", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="completion", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
litellm_debug_info=extra_information,
)
@ -1327,7 +1328,9 @@ def _map_vertex_exception(
response=httpx.Response(
status_code=500,
content=str(original_exception),
request=httpx.Request(method="completion", url="https://github.com/BerriAI/litellm"), # type: ignore
request=httpx.Request(
method="completion", url="https://github.com/BerriAI/litellm"
), # type: ignore
),
)
if original_exception.status_code == 502:
@ -1965,13 +1968,9 @@ def _map_azure_exception(
# content policy violation even when the top-level
# code is generic (e.g. "invalid_request_error").
if azure_error_code != "content_policy_violation":
_inner = body_dict["error"].get(
"inner_error"
) or body_dict[ # type: ignore[index]
_inner = body_dict["error"].get("inner_error") or body_dict[ # type: ignore[index]
"error"
].get(
"innererror"
) # type: ignore[index]
].get("innererror") # type: ignore[index]
if (
isinstance(_inner, dict)
and _inner.get("code") == "ResponsibleAIPolicyViolation"

10
litellm/litellm_core_utils/get_llm_provider_logic.py
View file

@ -616,7 +616,11 @@ def _get_openai_compatible_provider_info(
return model, "aiohttp_openai", api_key, api_base
elif custom_llm_provider == "anyscale":
# anyscale is openai compatible, we just need to set this to custom_openai and have the api_base be https://api.endpoints.anyscale.com/v1
api_base = api_base or get_secret_str("ANYSCALE_API_BASE") or "https://api.endpoints.anyscale.com/v1" # type: ignore
api_base = (
api_base
or get_secret_str("ANYSCALE_API_BASE")
or "https://api.endpoints.anyscale.com/v1"
) # type: ignore
dynamic_api_key = api_key or get_secret_str("ANYSCALE_API_KEY")
elif custom_llm_provider == "deepinfra":
(
@ -709,9 +713,7 @@ def _get_openai_compatible_provider_info(
) # type: ignore
dynamic_api_key = api_key or get_secret_str("NEBIUS_API_KEY")
elif custom_llm_provider == "ollama":
api_base = (
api_base or get_secret("OLLAMA_API_BASE") or "http://localhost:11434"
) # type: ignore
api_base = api_base or get_secret("OLLAMA_API_BASE") or "http://localhost:11434" # type: ignore
dynamic_api_key = api_key or get_secret_str("OLLAMA_API_KEY")
elif (custom_llm_provider == "ai21_chat") or (
custom_llm_provider == "ai21" and model in litellm.ai21_chat_models

88
litellm/litellm_core_utils/litellm_logging.py
View file

@ -293,7 +293,17 @@ def _get_cached_prometheus_logger():
class Logging(LiteLLMLoggingBaseClass):
global supabaseClient, promptLayerLogger, weightsBiasesLogger, logfireLogger, capture_exception, add_breadcrumb, lunaryLogger, logfireLogger, prometheusLogger, slack_app
global \
supabaseClient, \
promptLayerLogger, \
weightsBiasesLogger, \
logfireLogger, \
capture_exception, \
add_breadcrumb, \
lunaryLogger, \
logfireLogger, \
prometheusLogger, \
slack_app
custom_pricing: bool = False
stream_options = None
litellm_request_debug: bool = False
@ -359,9 +369,9 @@ class Logging(LiteLLMLoggingBaseClass):
)
self.function_id = function_id
self.streaming_chunks: List[Any] = [] # for generating complete stream response
self.sync_streaming_chunks: List[Any] = (
[]
) # for generating complete stream response
self.sync_streaming_chunks: List[
Any
] = [] # for generating complete stream response
self.log_raw_request_response = log_raw_request_response
# Initialize dynamic callbacks
@ -903,8 +913,11 @@ class Logging(LiteLLMLoggingBaseClass):
self.model_call_details["prompt_integration"] = logger.__class__.__name__
return logger
if anthropic_cache_control_logger := AnthropicCacheControlHook.get_custom_logger_for_anthropic_cache_control_hook(
non_default_params
if (
anthropic_cache_control_logger
:= AnthropicCacheControlHook.get_custom_logger_for_anthropic_cache_control_hook(
non_default_params
)
):
self.model_call_details["prompt_integration"] = (
anthropic_cache_control_logger.__class__.__name__
@ -978,9 +991,7 @@ class Logging(LiteLLMLoggingBaseClass):
self.model_call_details["api_key"] = api_key
self.model_call_details["additional_args"] = additional_args
self.model_call_details["log_event_type"] = "pre_api_call"
if (
model
): # if model name was changes pre-call, overwrite the initial model call name with the new one
if model: # if model name was changes pre-call, overwrite the initial model call name with the new one
self.model_call_details["model"] = model
self.model_call_details["litellm_params"]["api_base"] = (
self._get_masked_api_base(additional_args.get("api_base", ""))
@ -1359,13 +1370,13 @@ class Logging(LiteLLMLoggingBaseClass):
for callback in callbacks:
try:
if isinstance(callback, CustomLogger):
response: Optional[MCPPostCallResponseObject] = (
await callback.async_post_mcp_tool_call_hook(
kwargs=kwargs,
response_obj=post_mcp_tool_call_response_obj,
start_time=start_time,
end_time=end_time,
)
response: Optional[
MCPPostCallResponseObject
] = await callback.async_post_mcp_tool_call_hook(
kwargs=kwargs,
response_obj=post_mcp_tool_call_response_obj,
start_time=start_time,
end_time=end_time,
)
######################################################################
# if any of the callbacks modify the response, use the modified response
@ -1865,7 +1876,9 @@ class Logging(LiteLLMLoggingBaseClass):
self.model_call_details["litellm_params"].setdefault("metadata", {})
if self.model_call_details["litellm_params"]["metadata"] is None:
self.model_call_details["litellm_params"]["metadata"] = {}
self.model_call_details["litellm_params"]["metadata"]["hidden_params"] = getattr(logging_result, "_hidden_params", {}) # type: ignore
self.model_call_details["litellm_params"]["metadata"][
"hidden_params"
] = getattr(logging_result, "_hidden_params", {}) # type: ignore
if self.model_call_details.get("cache_hit") is True:
self.model_call_details["response_cost"] = 0.0
@ -1944,7 +1957,9 @@ class Logging(LiteLLMLoggingBaseClass):
)
result = result.model_copy()
transformed_usage = TranscriptionUsageObjectTransformation.transform_transcription_usage_object(result.usage) # type: ignore
transformed_usage = TranscriptionUsageObjectTransformation.transform_transcription_usage_object(
result.usage
) # type: ignore
setattr(result, "usage", transformed_usage)
return result
@ -2948,7 +2963,9 @@ class Logging(LiteLLMLoggingBaseClass):
for callback_obj in all_callbacks:
if hasattr(callback_obj, "increment_callback_logging_failure"):
callback_obj.increment_callback_logging_failure(callback_name=callback_name) # type: ignore
callback_obj.increment_callback_logging_failure(
callback_name=callback_name
) # type: ignore
break # Only increment once
except Exception as e:
@ -3292,9 +3309,7 @@ class Logging(LiteLLMLoggingBaseClass):
except Exception as e:
verbose_logger.exception(
"LiteLLM.LoggingError: [Non-Blocking] Exception occurred while failure \
logging {}\nCallback={}".format(
str(e), callback
)
logging {}\nCallback={}".format(str(e), callback)
)
# Track callback logging failures in Prometheus
self._handle_callback_failure(callback=callback)
@ -3762,7 +3777,29 @@ def set_callbacks(callback_list, function_id=None):
"""
Globally sets the callback client
"""
global sentry_sdk_instance, capture_exception, add_breadcrumb, slack_app, alerts_channel, traceloopLogger, athinaLogger, heliconeLogger, supabaseClient, lunaryLogger, promptLayerLogger, langFuseLogger, customLogger, weightsBiasesLogger, logfireLogger, dynamoLogger, s3Logger, dataDogLogger, prometheusLogger, greenscaleLogger, openMeterLogger, deepevalLogger
global \
sentry_sdk_instance, \
capture_exception, \
add_breadcrumb, \
slack_app, \
alerts_channel, \
traceloopLogger, \
athinaLogger, \
heliconeLogger, \
supabaseClient, \
lunaryLogger, \
promptLayerLogger, \
langFuseLogger, \
customLogger, \
weightsBiasesLogger, \
logfireLogger, \
dynamoLogger, \
s3Logger, \
dataDogLogger, \
prometheusLogger, \
greenscaleLogger, \
openMeterLogger, \
deepevalLogger
try:
for callback in callback_list:
@ -4607,7 +4644,7 @@ def _maybe_auto_initialize_arize_phoenix(_in_memory_loggers: list) -> None:
litellm.logging_callback_manager.add_litellm_callback(phoenix_logger)
verbose_logger.info(
"Auto-initialized Arize Phoenix logger alongside otel " "(endpoint=%s)",
"Auto-initialized Arize Phoenix logger alongside otel (endpoint=%s)",
arize_phoenix_config.endpoint,
)
except Exception as e:
@ -5781,7 +5818,8 @@ def get_standard_logging_object_payload(
id = f"{id}_cache_hit{time.time()}" # do not duplicate the request id
saved_cache_cost = (
logging_obj._response_cost_calculator(
result=init_response_obj, cache_hit=False # type: ignore
result=init_response_obj,
cache_hit=False, # type: ignore
)
or 0.0
)

12
litellm/litellm_core_utils/llm_cost_calc/utils.py
View file

@ -131,8 +131,10 @@ def _generic_cost_per_character(
assert (
"input_cost_per_character" in model_info
and model_info["input_cost_per_character"] is not None
), "model info for model={} does not have 'input_cost_per_character'-pricing\nmodel_info={}".format(
model, model_info
), (
"model info for model={} does not have 'input_cost_per_character'-pricing\nmodel_info={}".format(
model, model_info
)
)
custom_prompt_cost = model_info["input_cost_per_character"]
@ -152,8 +154,10 @@ def _generic_cost_per_character(
assert (
"output_cost_per_character" in model_info
and model_info["output_cost_per_character"] is not None
), "model info for model={} does not have 'output_cost_per_character'-pricing\nmodel_info={}".format(
model, model_info
), (
"model info for model={} does not have 'output_cost_per_character'-pricing\nmodel_info={}".format(
model, model_info
)
)
custom_completion_cost = model_info["output_cost_per_character"]
completion_cost = completion_characters * custom_completion_cost

24
litellm/litellm_core_utils/llm_response_utils/convert_dict_to_response.py
View file

@ -333,9 +333,15 @@ def convert_to_streaming_response(
if "usage" in response_object and response_object["usage"] is not None:
setattr(model_response_object, "usage", Usage())
model_response_object.usage.completion_tokens = response_object["usage"].get("completion_tokens", 0) # type: ignore
model_response_object.usage.prompt_tokens = response_object["usage"].get("prompt_tokens", 0) # type: ignore
model_response_object.usage.total_tokens = response_object["usage"].get("total_tokens", 0) # type: ignore
model_response_object.usage.completion_tokens = response_object["usage"].get(
"completion_tokens", 0
) # type: ignore
model_response_object.usage.prompt_tokens = response_object["usage"].get(
"prompt_tokens", 0
) # type: ignore
model_response_object.usage.total_tokens = response_object["usage"].get(
"total_tokens", 0
) # type: ignore
if "id" in response_object:
model_response_object.id = response_object["id"]
@ -848,9 +854,15 @@ def convert_to_model_response_object(
model_response_object.data = response_object["data"]
if "usage" in response_object and response_object["usage"] is not None:
model_response_object.usage.completion_tokens = response_object["usage"].get("completion_tokens", 0) # type: ignore
model_response_object.usage.prompt_tokens = response_object["usage"].get("prompt_tokens", 0) # type: ignore
model_response_object.usage.total_tokens = response_object["usage"].get("total_tokens", 0) # type: ignore
model_response_object.usage.completion_tokens = response_object[
"usage"
].get("completion_tokens", 0) # type: ignore
model_response_object.usage.prompt_tokens = response_object[
"usage"
].get("prompt_tokens", 0) # type: ignore
model_response_object.usage.total_tokens = response_object["usage"].get(
"total_tokens", 0
) # type: ignore
if start_time is not None and end_time is not None:
model_response_object._response_ms = ( # type: ignore

3
litellm/litellm_core_utils/logging_callback_manager.py
View file

@ -68,7 +68,8 @@ class LoggingCallbackManager:
Ensures no duplicates are added.
"""
self._safe_add_callback_to_list(
callback=callback, parent_list=litellm.callbacks # type: ignore
callback=callback,
parent_list=litellm.callbacks, # type: ignore
)
def add_litellm_success_callback(

14
litellm/litellm_core_utils/prompt_templates/factory.py
View file

@ -1586,7 +1586,9 @@ def convert_to_gemini_tool_call_result(
file_data = (
file_content.get("file_data", "")
if isinstance(file_content, dict)
else file_content if isinstance(file_content, str) else ""
else file_content
if isinstance(file_content, str)
else ""
)
if file_data:
@ -2556,9 +2558,7 @@ def anthropic_messages_pt(
ChatCompletionToolMessage,
ChatCompletionUserMessage,
ChatCompletionFunctionMessage,
] = messages[
msg_i
] # type: ignore
] = messages[msg_i] # type: ignore
if user_message_types_block["role"] == "user":
if isinstance(user_message_types_block["content"], list):
for m in user_message_types_block["content"]:
@ -4926,8 +4926,10 @@ class BedrockConverseMessagesProcessor:
image_url = element["image_url"]["url"]
else:
image_url = element["image_url"]
assistants_part = await BedrockImageProcessor.process_image_async( # type: ignore
image_url=image_url
assistants_part = (
await BedrockImageProcessor.process_image_async( # type: ignore
image_url=image_url
)
)
assistants_parts.append(assistants_part)
# Add cache point block for assistant content elements

4
litellm/litellm_core_utils/realtime_streaming.py
View file

@ -169,7 +169,9 @@ class RealTimeStreaming:
try:
event_type = message_obj.get("type", "")
if event_type in self._SESSION_EVENT_TYPES:
typed_obj: OpenAIRealtimeEvents = OpenAIRealtimeStreamSessionEvents(**message_obj) # type: ignore
typed_obj: OpenAIRealtimeEvents = OpenAIRealtimeStreamSessionEvents(
**message_obj
) # type: ignore
else:
# Catch-all base object so unknown/new event names never raise.
typed_obj = OpenAIRealtimeStreamResponseBaseObject(**message_obj) # type: ignore

16
litellm/litellm_core_utils/rules.py
View file

@ -33,7 +33,11 @@ class Rules:
if callable(rule):
decision = rule(input)
if decision is False:
raise litellm.APIResponseValidationError(message="LLM Response failed post-call-rule check", llm_provider="", model=model) # type: ignore
raise litellm.APIResponseValidationError(
message="LLM Response failed post-call-rule check",
llm_provider="",
model=model,
) # type: ignore
return True
def post_call_rules(self, input: Optional[str], model: str) -> bool:
@ -44,12 +48,18 @@ class Rules:
decision = rule(input)
if isinstance(decision, bool):
if decision is False:
raise litellm.APIResponseValidationError(message="LLM Response failed post-call-rule check", llm_provider="", model=model) # type: ignore
raise litellm.APIResponseValidationError(
message="LLM Response failed post-call-rule check",
llm_provider="",
model=model,
) # type: ignore
elif isinstance(decision, dict):
decision_val = decision.get("decision", True)
decision_message = decision.get(
"message", "LLM Response failed post-call-rule check"
)
if decision_val is False:
raise litellm.APIResponseValidationError(message=decision_message, llm_provider="", model=model) # type: ignore
raise litellm.APIResponseValidationError(
message=decision_message, llm_provider="", model=model
) # type: ignore
return True

4
litellm/litellm_core_utils/sensitive_data_masker.py
View file

@ -54,9 +54,9 @@ class SensitiveDataMasker:
# Handle the case where visible_suffix is 0 to avoid showing the entire string
if self.visible_suffix == 0:
return f"{value_str[:self.visible_prefix]}{self.mask_char * masked_length}"
return f"{value_str[: self.visible_prefix]}{self.mask_char * masked_length}"
else:
return f"{value_str[:self.visible_prefix]}{self.mask_char * masked_length}{value_str[-self.visible_suffix:]}"
return f"{value_str[: self.visible_prefix]}{self.mask_char * masked_length}{value_str[-self.visible_suffix :]}"
def is_sensitive_key(
self, key: str, excluded_keys: Optional[Set[str]] = None

21
litellm/litellm_core_utils/streaming_chunk_builder_utils.py
View file

@ -213,9 +213,9 @@ class ChunkProcessor:
self, tool_call_chunks: List[Dict[str, Any]]
) -> List[ChatCompletionMessageToolCall]:
tool_calls_list: List[ChatCompletionMessageToolCall] = []
tool_call_map: Dict[int, Dict[str, Any]] = (
{}
) # Map to store tool calls by index
tool_call_map: Dict[
int, Dict[str, Any]
] = {} # Map to store tool calls by index
for chunk in tool_call_chunks:
choices = chunk["choices"]
@ -720,15 +720,16 @@ class ChunkProcessor:
returned_usage.prompt_tokens = prompt_tokens or token_counter(
model=model, messages=messages
)
except (
Exception
): # don't allow this failing to block a complete streaming response from being returned
except Exception: # don't allow this failing to block a complete streaming response from being returned
print_verbose("token_counter failed, assuming prompt tokens is 0")
returned_usage.prompt_tokens = 0
returned_usage.completion_tokens = completion_tokens or token_counter(
model=model,
text=completion_output,
count_response_tokens=True, # count_response_tokens is a Flag to tell token counter this is a response, No need to add extra tokens we do for input messages
returned_usage.completion_tokens = (
completion_tokens
or token_counter(
model=model,
text=completion_output,
count_response_tokens=True, # count_response_tokens is a Flag to tell token counter this is a response, No need to add extra tokens we do for input messages
)
)
returned_usage.total_tokens = (
returned_usage.prompt_tokens + returned_usage.completion_tokens

18
litellm/litellm_core_utils/streaming_handler.py
View file

@ -189,9 +189,7 @@ class CustomStreamWrapper:
True if self.check_send_stream_usage(self.stream_options) else False
)
self.tool_call = False
self.chunks: List = (
[]
) # keep track of the returned chunks - used for calculating the input/output tokens for stream options
self.chunks: List = [] # keep track of the returned chunks - used for calculating the input/output tokens for stream options
self._repeated_messages_count = 1
self.is_function_call = self.check_is_function_call(logging_obj=logging_obj)
self.created: Optional[int] = None
@ -1861,8 +1859,10 @@ class CustomStreamWrapper:
Caches the streaming response
"""
if not cache_hit and self.logging_obj._llm_caching_handler is not None:
await self.logging_obj._llm_caching_handler._add_streaming_response_to_cache(
processed_chunk
await (
self.logging_obj._llm_caching_handler._add_streaming_response_to_cache(
processed_chunk
)
)
def run_success_logging_and_cache_storage(self, processed_chunk, cache_hit: bool):
@ -2217,7 +2217,9 @@ class CustomStreamWrapper:
)
)
# Add MCP metadata to final chunk if present (after hooks)
processed_chunk = self._add_mcp_metadata_to_final_chunk(processed_chunk) # type: ignore[reportArgumentType]
processed_chunk = self._add_mcp_metadata_to_final_chunk(
processed_chunk
) # type: ignore[reportArgumentType]
return processed_chunk
raise StopAsyncIteration
@ -2229,7 +2231,9 @@ class CustomStreamWrapper:
):
chunk = self.completion_stream
else:
chunk = await asyncio.to_thread(_next_sync_or_exhausted, self.completion_stream) # type: ignore[arg-type]
chunk = await asyncio.to_thread(
_next_sync_or_exhausted, self.completion_stream
) # type: ignore[arg-type]
if chunk is _SYNC_ITER_EXHAUSTED:
raise StopAsyncIteration
if chunk is not None and chunk != b"":

6
litellm/llms/__init__.py
View file

@ -63,9 +63,9 @@ def get_cost_for_web_search_request(
return None
def discover_guardrail_translation_mappings() -> (
Dict[CallTypes, Type["BaseTranslation"]]
):
def discover_guardrail_translation_mappings() -> Dict[
CallTypes, Type["BaseTranslation"]
]:
"""
Discover guardrail translation mappings by scanning the llms directory structure.

6
litellm/llms/anthropic/chat/guardrail_translation/handler.py
View file

@ -288,9 +288,9 @@ class AnthropicMessagesHandler(BaseTranslation):
elif isinstance(content, list) and content_idx_optional is not None:
# Replace specific text item in list content
messages[msg_idx]["content"][content_idx_optional][
"text"
] = guardrail_response
messages[msg_idx]["content"][content_idx_optional]["text"] = (
guardrail_response
)
async def process_output_response(
self,

4
litellm/llms/anthropic/chat/handler.py
View file

@ -624,7 +624,9 @@ class ModelResponseIterator:
speed=self.speed,
)
def _content_block_delta_helper(self, chunk: dict) -> Tuple[
def _content_block_delta_helper(
self, chunk: dict
) -> Tuple[
str,
Optional[ChatCompletionToolCallChunk],
List[Union[ChatCompletionThinkingBlock, ChatCompletionRedactedThinkingBlock]],

8
litellm/llms/anthropic/chat/transformation.py
View file

@ -750,7 +750,9 @@ class AnthropicConfig(AnthropicModelInfo, BaseConfig):
additional_tool_params[k] = v
returned_tool = AnthropicHostedTools(
type=tool["type"], name=function_name, **additional_tool_params # type: ignore
type=tool["type"],
name=function_name,
**additional_tool_params, # type: ignore
)
elif tool["type"] == "url": # mcp server tool
mcp_server = AnthropicMcpServerTool(**tool) # type: ignore
@ -2144,7 +2146,9 @@ class AnthropicConfig(AnthropicModelInfo, BaseConfig):
filtered_tools = [t for i, t in enumerate(tool_calls) if i not in json_indices]
return None, filtered_tools, extra_content
def extract_response_content(self, completion_response: dict) -> Tuple[
def extract_response_content(
self, completion_response: dict
) -> Tuple[
str,
Optional[List[Any]],
Optional[

3
litellm/llms/anthropic/common_utils.py
View file

@ -502,7 +502,8 @@ class AnthropicModelInfo(BaseLLMModelInfo):
"computer_20241022": "computer-use-2024-10-22",
}
return computer_tool_beta_mapping.get(
computer_tool_version, "computer-use-2024-10-22" # Default fallback
computer_tool_version,
"computer-use-2024-10-22", # Default fallback
)
def get_anthropic_beta_list(

4
litellm/llms/anthropic/experimental_pass_through/adapters/streaming_iterator.py
View file

@ -184,9 +184,7 @@ class AnthropicStreamWrapper(AdapterCompletionStreamWrapper):
# class level) so concurrent streams don't share the same mutable dict
# — `_should_start_new_content_block` mutates `tool_block["name"]` in
# place, which would otherwise leak across streams.
self.current_content_block_start: (
"AnthropicStreamWrapper.ContentBlockContentBlockDict"
) = self.TextBlock(
self.current_content_block_start: "AnthropicStreamWrapper.ContentBlockContentBlockDict" = self.TextBlock(
type="text",
text="",
)

18
litellm/llms/anthropic/experimental_pass_through/adapters/transformation.py
View file

@ -595,9 +595,9 @@ class LiteLLMAnthropicMessagesAdapter:
## ASSISTANT MESSAGE ##
assistant_message_str: Optional[str] = None
assistant_content_list: List[Dict[str, Any]] = (
[]
) # For content blocks with cache_control
assistant_content_list: List[
Dict[str, Any]
] = [] # For content blocks with cache_control
has_cache_control_in_text = False
tool_calls: List[ChatCompletionAssistantToolCall] = []
thinking_blocks: List[
@ -1025,7 +1025,9 @@ class LiteLLMAnthropicMessagesAdapter:
if openai_system_content:
new_messages.insert(
0,
ChatCompletionSystemMessage(role="system", content=openai_system_content), # type: ignore
ChatCompletionSystemMessage(
role="system", content=openai_system_content
), # type: ignore
)
def _translate_metadata_to_openai(
@ -1456,7 +1458,9 @@ class LiteLLMAnthropicMessagesAdapter:
"input_tokens": uncached_input_tokens,
"output_tokens": usage.completion_tokens or 0,
}
anthropic_usage["iterations"] = list(polyfill_result.iterations_usage) + [message_iteration] # type: ignore[typeddict-unknown-key]
anthropic_usage["iterations"] = list(polyfill_result.iterations_usage) + [
message_iteration
] # type: ignore[typeddict-unknown-key]
translated_obj = AnthropicMessagesResponse(
id=response.id,
@ -1672,7 +1676,9 @@ class LiteLLMAnthropicMessagesAdapter:
else:
usage_delta = UsageDelta(input_tokens=0, output_tokens=0)
message_block = MessageBlockDelta(
type="message_delta", delta=delta, usage=usage_delta # type: ignore
type="message_delta",
delta=delta,
usage=usage_delta, # type: ignore
)
if applied_edits:
message_block["context_management"] = ContextManagementResponse(

2
litellm/llms/anthropic/experimental_pass_through/messages/streaming_iterator.py
View file

@ -87,7 +87,7 @@ class BaseAnthropicMessagesStreamingIterator:
"""
if isinstance(chunk, dict):
event_type: str = str(chunk.get("type", "message"))
payload = f"event: {event_type}\n" f"data: {json.dumps(chunk)}\n\n"
payload = f"event: {event_type}\ndata: {json.dumps(chunk)}\n\n"
return payload.encode()
else:
# For non-dict chunks, return as is

6
litellm/llms/anthropic/experimental_pass_through/responses_adapters/streaming_iterator.py
View file

@ -35,9 +35,9 @@ class AnthropicResponsesStreamWrapper:
# Map item_id -> content_block_index so we can stop the right block later
self._item_id_to_block_index: Dict[str, int] = {}
# Track open function_call items by item_id so we can emit tool_use start
self._pending_tool_ids: Dict[str, str] = (
{}
) # item_id -> call_id / name accumulator
self._pending_tool_ids: Dict[
str, str
] = {} # item_id -> call_id / name accumulator
self._sent_message_start = False
self._sent_message_stop = False
self._chunk_queue: deque = deque()

10
litellm/llms/azure/assistants.py
View file

@ -203,8 +203,11 @@ class AzureAssistantsAPI(BaseAzureLLM):
litellm_params=litellm_params,
)
thread_message: OpenAIMessage = await openai_client.beta.threads.messages.create( # type: ignore
thread_id, **message_data # type: ignore
thread_message: OpenAIMessage = (
await openai_client.beta.threads.messages.create( # type: ignore
thread_id,
**message_data, # type: ignore
)
)
response_obj: Optional[OpenAIMessage] = None
@ -292,7 +295,8 @@ class AzureAssistantsAPI(BaseAzureLLM):
)
thread_message: OpenAIMessage = openai_client.beta.threads.messages.create( # type: ignore
thread_id, **message_data # type: ignore
thread_id,
**message_data, # type: ignore
)
response_obj: Optional[OpenAIMessage] = None

10
litellm/llms/azure/audio_transcriptions.py
View file

@ -79,7 +79,8 @@ class AzureAudioTranscription(AzureChatCompletion):
)
response = azure_client.audio.transcriptions.create(
**data, timeout=timeout # type: ignore
**data,
timeout=timeout, # type: ignore
)
if isinstance(response, BaseModel):
@ -95,7 +96,12 @@ class AzureAudioTranscription(AzureChatCompletion):
original_response=stringified_response,
)
hidden_params = {"model": model, "custom_llm_provider": "azure"}
final_response: TranscriptionResponse = convert_to_model_response_object(response_object=stringified_response, model_response_object=model_response, hidden_params=hidden_params, response_type="audio_transcription") # type: ignore
final_response: TranscriptionResponse = convert_to_model_response_object(
response_object=stringified_response,
model_response_object=model_response,
hidden_params=hidden_params,
response_type="audio_transcription",
) # type: ignore
return final_response
async def async_audio_transcriptions(

30
litellm/llms/azure/azure.py
View file

@ -817,7 +817,9 @@ class AzureChatCompletion(BaseAzureLLM, BaseLLM):
)
## COMPLETION CALL
raw_response = azure_client.embeddings.with_raw_response.create(**data, timeout=timeout) # type: ignore
raw_response = azure_client.embeddings.with_raw_response.create(
**data, timeout=timeout
) # type: ignore
headers = dict(raw_response.headers)
response = raw_response.parse()
if isinstance(response, str):
@ -833,7 +835,12 @@ class AzureChatCompletion(BaseAzureLLM, BaseLLM):
original_response=response,
)
return convert_to_model_response_object(response_object=response.model_dump(), model_response_object=model_response, response_type="embedding", _response_headers=process_azure_headers(headers)) # type: ignore
return convert_to_model_response_object(
response_object=response.model_dump(),
model_response_object=model_response,
response_type="embedding",
_response_headers=process_azure_headers(headers),
) # type: ignore
except AzureOpenAIError as e:
raise e
except Exception as e:
@ -1296,7 +1303,18 @@ class AzureChatCompletion(BaseAzureLLM, BaseLLM):
is_async=False,
)
if aimg_generation is True:
return self.aimage_generation(data=data, input=input, logging_obj=logging_obj, model_response=model_response, api_key=api_key, client=client, azure_client_params=azure_client_params, timeout=timeout, headers=headers, model=model) # type: ignore
return self.aimage_generation(
data=data,
input=input,
logging_obj=logging_obj,
model_response=model_response,
api_key=api_key,
client=client,
azure_client_params=azure_client_params,
timeout=timeout,
headers=headers,
model=model,
) # type: ignore
img_gen_api_base = self.create_azure_base_url(
azure_client_params=azure_client_params,
@ -1348,7 +1366,11 @@ class AzureChatCompletion(BaseAzureLLM, BaseLLM):
original_response=response,
)
# return response
return convert_to_model_response_object(response_object=response, model_response_object=model_response, response_type="image_generation") # type: ignore
return convert_to_model_response_object(
response_object=response,
model_response_object=model_response,
response_type="image_generation",
) # type: ignore
except AzureOpenAIError as e:
raise e
except Exception as e:

4
litellm/llms/azure/batches/handler.py
View file

@ -75,7 +75,9 @@ class AzureBatchesAPI(BaseAzureLLM):
return self.acreate_batch( # type: ignore
create_batch_data=create_batch_data, azure_client=azure_client
)
response = cast(Union[AzureOpenAI, OpenAI], azure_client).batches.create(**create_batch_data) # type: ignore[arg-type]
response = cast(Union[AzureOpenAI, OpenAI], azure_client).batches.create(
**create_batch_data
) # type: ignore[arg-type]
return LiteLLMBatch(**response.model_dump())
async def aretrieve_batch(

8
litellm/llms/azure/files/handler.py
View file

@ -45,7 +45,9 @@ class AzureOpenAIFilesAPI(BaseAzureLLM):
openai_client: Union[AsyncAzureOpenAI, AsyncOpenAI],
) -> OpenAIFileObject:
verbose_logger.debug("create_file_data=%s", create_file_data)
response = await openai_client.files.create(**self._prepare_create_file_data(create_file_data)) # type: ignore[arg-type]
response = await openai_client.files.create(
**self._prepare_create_file_data(create_file_data)
) # type: ignore[arg-type]
verbose_logger.debug("create_file_response=%s", response)
return OpenAIFileObject(**response.model_dump())
@ -86,7 +88,9 @@ class AzureOpenAIFilesAPI(BaseAzureLLM):
return self.acreate_file(
create_file_data=create_file_data, openai_client=openai_client
)
response = cast(Union[AzureOpenAI, OpenAI], openai_client).files.create(**self._prepare_create_file_data(create_file_data)) # type: ignore[arg-type]
response = cast(Union[AzureOpenAI, OpenAI], openai_client).files.create(
**self._prepare_create_file_data(create_file_data)
) # type: ignore[arg-type]
return OpenAIFileObject(**response.model_dump())
async def afile_content(

9
litellm/llms/bedrock/chat/agentcore/transformation.py
View file

@ -403,8 +403,7 @@ class AmazonAgentCoreConfig(BaseConfig, BaseAWSLLM):
# skip strategy matching and fall back to raw JSON string
if not isinstance(response_json, dict):
verbose_logger.warning(
"AgentCore: JSON response is not a dict. "
"Returning raw JSON as content."
"AgentCore: JSON response is not a dict. Returning raw JSON as content."
)
return AgentCoreParsedResponse(
content=json.dumps(response_json),
@ -940,9 +939,9 @@ class AmazonAgentCoreConfig(BaseConfig, BaseAWSLLM):
)
parsed = self._parse_json_response(response_json)
async def _json_as_async_stream() -> (
AsyncGenerator[ModelResponseStream, None]
):
async def _json_as_async_stream() -> AsyncGenerator[
ModelResponseStream, None
]:
# Content chunk
content_chunk = ModelResponseStream(
id=f"chatcmpl-{uuid.uuid4()}",

26
litellm/llms/bedrock/chat/converse_handler.py
View file

@ -51,19 +51,19 @@ def make_sync_call(
)
if fake_stream:
model_response: (
ModelResponse
) = litellm.AmazonConverseConfig()._transform_response(
model=model,
response=response,
model_response=litellm.ModelResponse(),
stream=True,
logging_obj=logging_obj,
optional_params={},
api_key="",
data=data,
messages=messages,
encoding=litellm.encoding,
model_response: ModelResponse = (
litellm.AmazonConverseConfig()._transform_response(
model=model,
response=response,
model_response=litellm.ModelResponse(),
stream=True,
logging_obj=logging_obj,
optional_params={},
api_key="",
data=data,
messages=messages,
encoding=litellm.encoding,
)
) # type: ignore
completion_stream: Any = MockResponseIterator(
model_response=model_response, json_mode=json_mode

4
litellm/llms/bedrock/chat/converse_transformation.py
View file

@ -1963,7 +1963,9 @@ class AmazonConverseConfig(BaseConfig):
return message, returned_finish_reason
def _translate_message_content(self, content_blocks: List[ContentBlock]) -> Tuple[
def _translate_message_content(
self, content_blocks: List[ContentBlock]
) -> Tuple[
str,
List[ChatCompletionToolCallChunk],
Optional[List[BedrockConverseReasoningContentBlock]],

56
litellm/llms/bedrock/chat/invoke_handler.py
View file

@ -225,19 +225,19 @@ async def make_call(
raise BedrockError(status_code=response.status_code, message=response.text)
if fake_stream:
model_response: (
ModelResponse
) = litellm.AmazonConverseConfig()._transform_response(
model=model,
response=response,
model_response=litellm.ModelResponse(),
stream=True,
logging_obj=logging_obj,
optional_params={},
api_key="",
data=data,
messages=messages,
encoding=litellm.encoding,
model_response: ModelResponse = (
litellm.AmazonConverseConfig()._transform_response(
model=model,
response=response,
model_response=litellm.ModelResponse(),
stream=True,
logging_obj=logging_obj,
optional_params={},
api_key="",
data=data,
messages=messages,
encoding=litellm.encoding,
)
) # type: ignore
completion_stream: Any = MockResponseIterator(
model_response=model_response, json_mode=json_mode
@ -321,19 +321,19 @@ def make_sync_call(
raise BedrockError(status_code=response.status_code, message=response.text)
if fake_stream:
model_response: (
ModelResponse
) = litellm.AmazonConverseConfig()._transform_response(
model=model,
response=response,
model_response=litellm.ModelResponse(),
stream=True,
logging_obj=logging_obj,
optional_params={},
api_key="",
data=data,
messages=messages,
encoding=litellm.encoding,
model_response: ModelResponse = (
litellm.AmazonConverseConfig()._transform_response(
model=model,
response=response,
model_response=litellm.ModelResponse(),
stream=True,
logging_obj=logging_obj,
optional_params={},
api_key="",
data=data,
messages=messages,
encoding=litellm.encoding,
)
) # type: ignore
completion_stream: Any = MockResponseIterator(
model_response=model_response, json_mode=json_mode
@ -1300,7 +1300,9 @@ class BedrockLLM(BaseAWSLLM):
if isinstance(timeout, float) or isinstance(timeout, int):
timeout = httpx.Timeout(timeout)
_params["timeout"] = timeout
client = get_async_httpx_client(params=_params, llm_provider=litellm.LlmProviders.BEDROCK) # type: ignore
client = get_async_httpx_client(
params=_params, llm_provider=litellm.LlmProviders.BEDROCK
) # type: ignore
else:
client = client # type: ignore

12
litellm/llms/bedrock/embed/embedding.py
View file

@ -142,7 +142,9 @@ class BedrockEmbedding(BaseAWSLLM):
client = client
try:
response = await client.post(url=api_base, headers=headers, data=json.dumps(data)) # type: ignore
response = await client.post(
url=api_base, headers=headers, data=json.dumps(data)
) # type: ignore
response.raise_for_status()
except httpx.HTTPStatusError as err:
error_code = err.response.status_code
@ -451,10 +453,10 @@ class BedrockEmbedding(BaseAWSLLM):
batch_data = []
for i in input:
if model == "amazon.titan-embed-image-v1":
transformed_request: (
AmazonEmbeddingRequest
) = AmazonTitanMultimodalEmbeddingG1Config()._transform_request(
input=i, inference_params=inference_params
transformed_request: AmazonEmbeddingRequest = (
AmazonTitanMultimodalEmbeddingG1Config()._transform_request(
input=i, inference_params=inference_params
)
)
elif model == "amazon.titan-embed-text-v1":
transformed_request = AmazonTitanG1Config()._transform_request(

12
litellm/llms/bedrock/image_edit/handler.py
View file

@ -112,7 +112,11 @@ class BedrockImageEdit(BaseAWSLLM):
if client is None or not isinstance(client, HTTPHandler):
client = _get_httpx_client()
try:
response = client.post(url=prepared_request.endpoint_url, headers=prepared_request.prepped.headers, data=prepared_request.body) # type: ignore
response = client.post(
url=prepared_request.endpoint_url,
headers=prepared_request.prepped.headers,
data=prepared_request.body,
) # type: ignore
response.raise_for_status()
except httpx.HTTPStatusError as err:
error_code = err.response.status_code
@ -150,7 +154,11 @@ class BedrockImageEdit(BaseAWSLLM):
)
try:
response = await async_client.post(url=prepared_request.endpoint_url, headers=prepared_request.prepped.headers, data=prepared_request.body) # type: ignore
response = await async_client.post(
url=prepared_request.endpoint_url,
headers=prepared_request.prepped.headers,
data=prepared_request.body,
) # type: ignore
response.raise_for_status()
except httpx.HTTPStatusError as err:
error_code = err.response.status_code

4
litellm/llms/bedrock/image_edit/stability_transformation.py
View file

@ -122,7 +122,9 @@ class BedrockStabilityImageEditConfig(BaseImageEditConfig):
if k in param_mapping:
# Map param if mapping exists and value is valid
if k == "size" and v in OPENAI_SIZE_TO_STABILITY_ASPECT_RATIO:
mapped_params[param_mapping[k]] = OPENAI_SIZE_TO_STABILITY_ASPECT_RATIO[v] # type: ignore
mapped_params[param_mapping[k]] = (
OPENAI_SIZE_TO_STABILITY_ASPECT_RATIO[v]
) # type: ignore
# Don't copy "size" itself to final dict
elif k == "n":
# Store for logic but do not add to outgoing params

11
litellm/llms/bedrock/image_generation/amazon_nova_canvas_transformation.py
View file

@ -111,8 +111,10 @@ class AmazonNovaCanvasConfig:
**color_guided_generation_params,
}
try:
color_guided_generation_params_typed = AmazonNovaCanvasColorGuidedGenerationParams(
**color_guided_generation_params # type: ignore
color_guided_generation_params_typed = (
AmazonNovaCanvasColorGuidedGenerationParams(
**color_guided_generation_params # type: ignore
)
)
except Exception as e:
raise ValueError(
@ -171,8 +173,9 @@ class AmazonNovaCanvasConfig:
_size = non_default_params.get("size")
if _size is not None:
width, height = _size.split("x")
optional_params["width"], optional_params["height"] = int(width), int(
height
optional_params["width"], optional_params["height"] = (
int(width),
int(height),
)
if non_default_params.get("n") is not None:
optional_params["numberOfImages"] = non_default_params.get("n")

12
litellm/llms/bedrock/image_generation/image_handler.py
View file

@ -115,7 +115,11 @@ class BedrockImageGeneration(BaseAWSLLM):
if client is None or not isinstance(client, HTTPHandler):
client = _get_httpx_client()
try:
response = client.post(url=prepared_request.endpoint_url, headers=prepared_request.prepped.headers, data=prepared_request.body) # type: ignore
response = client.post(
url=prepared_request.endpoint_url,
headers=prepared_request.prepped.headers,
data=prepared_request.body,
) # type: ignore
response.raise_for_status()
except httpx.HTTPStatusError as err:
error_code = err.response.status_code
@ -154,7 +158,11 @@ class BedrockImageGeneration(BaseAWSLLM):
)
try:
response = await async_client.post(url=prepared_request.endpoint_url, headers=prepared_request.prepped.headers, data=prepared_request.body) # type: ignore
response = await async_client.post(
url=prepared_request.endpoint_url,
headers=prepared_request.prepped.headers,
data=prepared_request.body,
) # type: ignore
response.raise_for_status()
except httpx.HTTPStatusError as err:
error_code = err.response.status_code

4
litellm/llms/bedrock/realtime/handler.py
View file

@ -237,9 +237,7 @@ class BedrockRealtime(BaseAWSLLM):
# Transform Bedrock format to OpenAI format
from litellm.types.realtime import RealtimeResponseTransformInput
realtime_response_transform_input: (
RealtimeResponseTransformInput
) = {
realtime_response_transform_input: RealtimeResponseTransformInput = {
"current_output_item_id": session_state.get(
"current_output_item_id"
),

8
litellm/llms/bedrock/rerank/handler.py
View file

@ -96,7 +96,13 @@ class BedrockRerankHandler(BaseAWSLLM):
)
if _is_async:
return self.arerank(prepared_request, timeout=timeout, client=client if client is not None and isinstance(client, AsyncHTTPHandler) else None) # type: ignore
return self.arerank(
prepared_request,
timeout=timeout,
client=client
if client is not None and isinstance(client, AsyncHTTPHandler)
else None,
) # type: ignore
if client is None or not isinstance(client, HTTPHandler):
client = _get_httpx_client()

6
litellm/llms/bedrock/vector_stores/transformation.py
View file

@ -253,9 +253,9 @@ class BedrockVectorStoreConfig(BaseVectorStoreConfig, BaseAWSLLM):
verbose_logger.debug(
"Overriding extra_body retrievalConfiguration.vectorSearchConfiguration.filter with filters from vector_store_search_optional_params"
)
retrieval_config.setdefault("vectorSearchConfiguration", {})[
"filter"
] = filters
retrieval_config.setdefault("vectorSearchConfiguration", {})["filter"] = (
filters
)
if retrieval_config:
request_body["retrievalConfiguration"] = cast(
BedrockKBRetrievalConfiguration, retrieval_config

101
litellm/llms/custom_httpx/http_handler.py
View file

@ -72,9 +72,9 @@ _AIOHTTP_SUPPORTS_SOCKET_FACTORY = (
)
def _build_aiohttp_keepalive_socket_factory() -> (
Optional[Callable[[Tuple[Any, ...]], socket.socket]]
):
def _build_aiohttp_keepalive_socket_factory() -> Optional[
Callable[[Tuple[Any, ...]], socket.socket]
]:
"""
Build a socket_factory that enables SO_KEEPALIVE on aiohttp TCP sockets.
@ -719,7 +719,14 @@ class AsyncHTTPHandler:
)
req = self.client.build_request(
"PUT", url, data=request_data, json=json, params=params, headers=headers, timeout=timeout, content=request_content # type: ignore
"PUT",
url,
data=request_data,
json=json,
params=params,
headers=headers,
timeout=timeout,
content=request_content, # type: ignore
)
response = await self.client.send(req)
response.raise_for_status()
@ -780,7 +787,14 @@ class AsyncHTTPHandler:
)
req = self.client.build_request(
"PATCH", url, data=request_data, json=json, params=params, headers=headers, timeout=timeout, content=request_content # type: ignore
"PATCH",
url,
data=request_data,
json=json,
params=params,
headers=headers,
timeout=timeout,
content=request_content, # type: ignore
)
response = await self.client.send(req)
response.raise_for_status()
@ -841,7 +855,14 @@ class AsyncHTTPHandler:
)
req = self.client.build_request(
"DELETE", url, data=request_data, json=json, params=params, headers=headers, timeout=timeout, content=request_content # type: ignore
"DELETE",
url,
data=request_data,
json=json,
params=params,
headers=headers,
timeout=timeout,
content=request_content, # type: ignore
)
response = await self.client.send(req, stream=stream)
response.raise_for_status()
@ -888,7 +909,13 @@ class AsyncHTTPHandler:
request_data, request_content = _prepare_request_data_and_content(data, content)
req = client.build_request(
"POST", url, data=request_data, json=json, params=params, headers=headers, content=request_content # type: ignore
"POST",
url,
data=request_data,
json=json,
params=params,
headers=headers,
content=request_content, # type: ignore
)
response = await client.send(req, stream=stream)
response.raise_for_status()
@ -1200,7 +1227,14 @@ class HTTPHandler:
)
else:
req = self.client.build_request(
"POST", url, data=request_data, json=json, params=params, headers=headers, files=files, content=request_content # type: ignore
"POST",
url,
data=request_data,
json=json,
params=params,
headers=headers,
files=files,
content=request_content, # type: ignore
)
response = self.client.send(req, stream=stream)
response.raise_for_status()
@ -1235,11 +1269,24 @@ class HTTPHandler:
if timeout is not None:
req = self.client.build_request(
"PATCH", url, data=request_data, json=json, params=params, headers=headers, timeout=timeout, content=request_content # type: ignore
"PATCH",
url,
data=request_data,
json=json,
params=params,
headers=headers,
timeout=timeout,
content=request_content, # type: ignore
)
else:
req = self.client.build_request(
"PATCH", url, data=request_data, json=json, params=params, headers=headers, content=request_content # type: ignore
"PATCH",
url,
data=request_data,
json=json,
params=params,
headers=headers,
content=request_content, # type: ignore
)
response = self.client.send(req, stream=stream)
response.raise_for_status()
@ -1274,11 +1321,24 @@ class HTTPHandler:
if timeout is not None:
req = self.client.build_request(
"PUT", url, data=request_data, json=json, params=params, headers=headers, timeout=timeout, content=request_content # type: ignore
"PUT",
url,
data=request_data,
json=json,
params=params,
headers=headers,
timeout=timeout,
content=request_content, # type: ignore
)
else:
req = self.client.build_request(
"PUT", url, data=request_data, json=json, params=params, headers=headers, content=request_content # type: ignore
"PUT",
url,
data=request_data,
json=json,
params=params,
headers=headers,
content=request_content, # type: ignore
)
response = self.client.send(req, stream=stream)
return response
@ -1312,11 +1372,24 @@ class HTTPHandler:
if timeout is not None:
req = self.client.build_request(
"DELETE", url, data=request_data, json=json, params=params, headers=headers, timeout=timeout, content=request_content # type: ignore
"DELETE",
url,
data=request_data,
json=json,
params=params,
headers=headers,
timeout=timeout,
content=request_content, # type: ignore
)
else:
req = self.client.build_request(
"DELETE", url, data=request_data, json=json, params=params, headers=headers, content=request_content # type: ignore
"DELETE",
url,
data=request_data,
json=json,
params=params,
headers=headers,
content=request_content, # type: ignore
)
response = self.client.send(req, stream=stream)
response.raise_for_status()

5
litellm/llms/custom_httpx/httpx_handler.py
View file

@ -54,7 +54,10 @@ class HTTPHandler:
):
try:
response = await self.client.post(
url, data=data, params=params, headers=headers # type: ignore
url,
data=data,
params=params,
headers=headers, # type: ignore
)
return response
except Exception as e:

10
litellm/llms/fireworks_ai/chat/transformation.py
View file

@ -578,11 +578,11 @@ class FireworksAIConfig(OpenAIGPTConfig):
## FIREWORKS AI sends tool calls in the content field instead of tool_calls
for choice in response.choices:
cast(Choices, choice).message = (
self._handle_message_content_with_tool_calls(
message=cast(Choices, choice).message,
tool_calls=optional_params.get("tools", None),
)
cast(
Choices, choice
).message = self._handle_message_content_with_tool_calls(
message=cast(Choices, choice).message,
tool_calls=optional_params.get("tools", None),
)
response._hidden_params = {

6
litellm/llms/gemini/realtime/transformation.py
View file

@ -1755,9 +1755,9 @@ class GeminiRealtimeConfig(BaseRealtimeConfig):
)
tool_call_temperature = tool_call_generation_config.get("temperature")
if tool_call_temperature is not None:
tool_call_done_event["response"][
"temperature"
] = tool_call_temperature
tool_call_done_event["response"]["temperature"] = (
tool_call_temperature
)
tool_call_max_output_tokens = tool_call_generation_config.get(
"maxOutputTokens"
)

4
litellm/llms/github_copilot/authenticator.py
View file

@ -182,7 +182,7 @@ class Authenticator:
)
except httpx.HTTPStatusError as e:
verbose_logger.error(
f"HTTP error refreshing API key (attempt {attempt+1}/{max_retries}): {str(e)}"
f"HTTP error refreshing API key (attempt {attempt + 1}/{max_retries}): {str(e)}"
)
except Exception as e:
verbose_logger.error(f"Unexpected error refreshing API key: {str(e)}")
@ -318,7 +318,7 @@ class Authenticator:
and resp_json.get("error") == "authorization_pending"
):
verbose_logger.debug(
f"Authorization pending (attempt {attempt+1}/{max_attempts})"
f"Authorization pending (attempt {attempt + 1}/{max_attempts})"
)
else:
verbose_logger.warning(f"Unexpected response: {resp_json}")

3
litellm/llms/github_copilot/responses/transformation.py
View file

@ -58,8 +58,7 @@ def github_copilot_supports_responses_api(model: str) -> bool:
)
except Exception as e:
verbose_logger.debug(
"github_copilot_supports_responses_api: get_model_info failed "
"for %s: %s",
"github_copilot_supports_responses_api: get_model_info failed for %s: %s",
model,
e,
)

6
litellm/llms/inception/chat/transformation.py
View file

@ -45,7 +45,11 @@ class InceptionChatConfig(OpenAILikeChatConfig):
self, api_base: Optional[str], api_key: Optional[str]
) -> Tuple[Optional[str], Optional[str]]:
passed_api_base = api_base
api_base = api_base or get_secret_str("INCEPTION_API_BASE") or "https://api.inceptionlabs.ai/v1" # type: ignore
api_base = (
api_base
or get_secret_str("INCEPTION_API_BASE")
or "https://api.inceptionlabs.ai/v1"
) # type: ignore
dynamic_api_key = api_key
if passed_api_base is None or api_key:
dynamic_api_key = (

6
litellm/llms/llamafile/chat/transformation.py
View file

@ -27,7 +27,11 @@ class LlamafileChatConfig(OpenAIGPTConfig):
If both are None, a default Llamafile server URL is returned.
See: https://github.com/Mozilla-Ocho/llamafile/blob/bd1bbe9aabb1ee12dbdcafa8936db443c571eb9d/README.md#L61
"""
return api_base or get_secret_str("LLAMAFILE_API_BASE") or "http://127.0.0.1:8080/v1" # type: ignore
return (
api_base
or get_secret_str("LLAMAFILE_API_BASE")
or "http://127.0.0.1:8080/v1"
) # type: ignore
def _get_openai_compatible_provider_info(
self, api_base: Optional[str], api_key: Optional[str]

6
litellm/llms/moonshot/chat/transformation.py
View file

@ -64,7 +64,11 @@ class MoonshotChatConfig(OpenAIGPTConfig):
def _get_openai_compatible_provider_info(
self, api_base: Optional[str], api_key: Optional[str]
) -> Tuple[Optional[str], Optional[str]]:
api_base = api_base or get_secret_str("MOONSHOT_API_BASE") or "https://api.moonshot.ai/v1" # type: ignore
api_base = (
api_base
or get_secret_str("MOONSHOT_API_BASE")
or "https://api.moonshot.ai/v1"
) # type: ignore
dynamic_api_key = api_key or get_secret_str("MOONSHOT_API_KEY")
return api_base, dynamic_api_key

10
litellm/llms/nvidia_nim/rerank/transformation.py
View file

@ -234,12 +234,18 @@ class NvidiaNimRerankConfig(BaseRerankConfig):
}
# Add optional top_k parameter if provided (already mapped from top_n in map_cohere_rerank_params)
if "top_k" in optional_rerank_params and optional_rerank_params.get("top_k") is not None: # type: ignore
if (
"top_k" in optional_rerank_params
and optional_rerank_params.get("top_k") is not None
): # type: ignore
request_data["top_k"] = optional_rerank_params.get("top_k") # type: ignore
# Add Nvidia-specific truncate parameter if provided
# This is passed through from non_default_params, not in base OptionalRerankParams
if "truncate" in optional_rerank_params and optional_rerank_params.get("truncate") is not None: # type: ignore
if (
"truncate" in optional_rerank_params
and optional_rerank_params.get("truncate") is not None
): # type: ignore
truncate_value = optional_rerank_params.get("truncate") # type: ignore
if truncate_value in ["NONE", "END"]:
request_data["truncate"] = truncate_value # type: ignore

4
litellm/llms/oci/chat/cohere.py
View file

@ -274,7 +274,9 @@ def handle_cohere_response(
total_tokens=usage_info.totalTokens,
)
else:
model_response.usage = Usage(prompt_tokens=0, completion_tokens=0, total_tokens=0) # type: ignore[attr-defined]
model_response.usage = Usage(
prompt_tokens=0, completion_tokens=0, total_tokens=0
) # type: ignore[attr-defined]
return model_response

3
litellm/llms/oci/chat/transformation.py
View file

@ -529,7 +529,8 @@ class OCIChatConfig(BaseConfig):
)
else:
selected_params["tools"] = adapt_tool_definition_to_oci_standard( # type: ignore[assignment]
selected_params["tools"], vendor # type: ignore[arg-type]
selected_params["tools"],
vendor, # type: ignore[arg-type]
)
# Normalise tool_choice to OCI's flat uppercase dict form

4
litellm/llms/oci/common_utils.py
View file

@ -339,7 +339,9 @@ def sign_with_manual_credentials(
private_key = (
load_private_key_from_str(oci_key_content)
if oci_key_content
else load_private_key_from_file(oci_key_file) if oci_key_file else None
else load_private_key_from_file(oci_key_file)
if oci_key_file
else None
)
if private_key is None:

4
litellm/llms/ollama/chat/transformation.py
View file

@ -411,7 +411,9 @@ class OllamaChatConfig(BaseConfig):
model_response.choices[0].finish_reason = "tool_calls"
model_response.created = int(time.time())
model_response.model = "ollama_chat/" + model
prompt_tokens = response_json.get("prompt_eval_count", litellm.token_counter(messages=messages)) # type: ignore
prompt_tokens = response_json.get(
"prompt_eval_count", litellm.token_counter(messages=messages)
) # type: ignore
completion_tokens = response_json.get(
"eval_count",
litellm.token_counter(text=response_json["message"]["content"]),

3
litellm/llms/ollama/completion/transformation.py
View file

@ -337,7 +337,8 @@ class OllamaConfig(BaseConfig):
model_response.model = "ollama/" + model
_prompt = request_data.get("prompt", "")
prompt_tokens = response_json.get(
"prompt_eval_count", len(encoding.encode(_prompt, disallowed_special=())) # type: ignore
"prompt_eval_count",
len(encoding.encode(_prompt, disallowed_special=())), # type: ignore
)
completion_tokens = response_json.get(
"eval_count", len(response_json.get("message", dict()).get("content", ""))

4
litellm/llms/oobabooga/chat/transformation.py
View file

@ -65,7 +65,9 @@ class OobaboogaConfig(OpenAIGPTConfig):
)
else:
try:
model_response.choices[0].message.content = completion_response["choices"][0]["message"]["content"] # type: ignore
model_response.choices[0].message.content = completion_response[
"choices"
][0]["message"]["content"] # type: ignore
except Exception as e:
raise OobaboogaError(
message=str(e),

11
litellm/llms/openai/chat/gpt_transformation.py
View file

@ -379,10 +379,10 @@ class OpenAIGPTConfig(BaseLLMModelInfo, BaseConfig):
List[OpenAIMessageContentListBlock], message_content
)
for i, content_item in enumerate(message_content_types):
message_content_types[i] = (
await self._async_transform_content_item(
cast(OpenAIMessageContentListBlock, content_item),
)
message_content_types[
i
] = await self._async_transform_content_item(
cast(OpenAIMessageContentListBlock, content_item),
)
return messages
@ -419,7 +419,8 @@ class OpenAIGPTConfig(BaseLLMModelInfo, BaseConfig):
for i, message in enumerate(messages):
messages[i] = cast(
AllMessageValues, filter_value_from_dict(message, "cache_control") # type: ignore
AllMessageValues,
filter_value_from_dict(message, "cache_control"), # type: ignore
)
if tools is not None:
for i, tool in enumerate(tools):

10
litellm/llms/openai/chat/guardrail_translation/handler.py
View file

@ -259,9 +259,9 @@ class OpenAIChatCompletionsHandler(BaseTranslation):
elif isinstance(content, list) and content_idx_optional is not None:
# Replace specific text item in list content
messages[msg_idx]["content"][content_idx_optional][
"text"
] = guardrail_response
messages[msg_idx]["content"][content_idx_optional]["text"] = (
guardrail_response
)
async def _apply_guardrail_responses_to_input_tool_calls(
self,
@ -746,7 +746,9 @@ class OpenAIChatCompletionsHandler(BaseTranslation):
elif isinstance(content, list) and content_idx_optional is not None:
# Replace specific text item in list content
choice.message.content[content_idx_optional]["text"] = guardrail_response # type: ignore
choice.message.content[content_idx_optional]["text"] = (
guardrail_response # type: ignore
)
async def _apply_guardrail_responses_to_output_tool_calls(
self,

18
litellm/llms/openai/completion/handler.py
View file

@ -96,7 +96,19 @@ class OpenAITextCompletion(BaseLLM):
organization=organization,
)
else:
return self.acompletion(api_base=api_base, data=data, headers=headers, model_response=model_response, api_key=api_key, logging_obj=logging_obj, model=model, timeout=timeout, max_retries=max_retries, organization=organization, client=client) # type: ignore
return self.acompletion(
api_base=api_base,
data=data,
headers=headers,
model_response=model_response,
api_key=api_key,
logging_obj=logging_obj,
model=model,
timeout=timeout,
max_retries=max_retries,
organization=organization,
client=client,
) # type: ignore
elif optional_params.get("stream", False):
return self.streaming(
logging_obj=logging_obj,
@ -124,7 +136,9 @@ class OpenAITextCompletion(BaseLLM):
else:
openai_client = client
raw_response = openai_client.completions.with_raw_response.create(**data) # type: ignore
raw_response = openai_client.completions.with_raw_response.create(
**data
) # type: ignore
response = raw_response.parse()
response_json = response.model_dump()

4
litellm/llms/openai/image_variations/handler.py
View file

@ -73,7 +73,9 @@ class OpenAIImageVariationsHandler:
client=client, init_client_params=init_client_params
)
raw_response = await client.images.with_raw_response.create_variation(**data) # type: ignore
raw_response = await client.images.with_raw_response.create_variation(
**data
) # type: ignore
response = raw_response.parse()
response_json = response.model_dump()

40
litellm/llms/openai/openai.py
View file

@ -1132,9 +1132,7 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
data = drop_params_from_unprocessable_entity_error(e, data)
else:
raise e
except (
Exception
) as e: # need to exception handle here. async exceptions don't get caught in sync functions.
except Exception as e: # need to exception handle here. async exceptions don't get caught in sync functions.
if isinstance(e, OpenAIError):
raise e
@ -1433,7 +1431,11 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
additional_args={"complete_input_dict": data},
original_response=stringified_response,
)
return convert_to_model_response_object(response_object=stringified_response, model_response_object=model_response, response_type="image_generation") # type: ignore
return convert_to_model_response_object(
response_object=stringified_response,
model_response_object=model_response,
response_type="image_generation",
) # type: ignore
except Exception as e:
## LOGGING
logging_obj.post_call(
@ -1466,7 +1468,19 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
raise OpenAIError(status_code=422, message="max retries must be an int")
if aimg_generation is True:
return self.aimage_generation(data=data, prompt=prompt, logging_obj=logging_obj, model_response=model_response, api_base=api_base, api_key=api_key, timeout=timeout, client=client, max_retries=max_retries, organization=organization, headers=headers) # type: ignore
return self.aimage_generation(
data=data,
prompt=prompt,
logging_obj=logging_obj,
model_response=model_response,
api_base=api_base,
api_key=api_key,
timeout=timeout,
client=client,
max_retries=max_retries,
organization=organization,
headers=headers,
) # type: ignore
openai_client: OpenAI = self._get_openai_client( # type: ignore
is_async=False,
@ -1503,7 +1517,11 @@ class OpenAIChatCompletion(BaseLLM, BaseOpenAILLM):
additional_args={"complete_input_dict": data},
original_response=response,
)
return convert_to_model_response_object(response_object=response, model_response_object=model_response, response_type="image_generation") # type: ignore
return convert_to_model_response_object(
response_object=response,
model_response_object=model_response,
response_type="image_generation",
) # type: ignore
except OpenAIError as e:
## LOGGING
logging_obj.post_call(
@ -2517,8 +2535,11 @@ class OpenAIAssistantsAPI(BaseLLM):
client=client,
)
thread_message: OpenAIMessage = await openai_client.beta.threads.messages.create( # type: ignore
thread_id, **message_data # type: ignore
thread_message: OpenAIMessage = (
await openai_client.beta.threads.messages.create( # type: ignore
thread_id,
**message_data, # type: ignore
)
)
response_obj: Optional[OpenAIMessage] = None
@ -2596,7 +2617,8 @@ class OpenAIAssistantsAPI(BaseLLM):
)
thread_message: OpenAIMessage = openai_client.beta.threads.messages.create( # type: ignore
thread_id, **message_data # type: ignore
thread_id,
**message_data, # type: ignore
)
response_obj: Optional[OpenAIMessage] = None

Some files were not shown because too many files have changed in this diff Show more