phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 02:34:29 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Gemini-2.5-flash - support reasoning cost calc + return reasoning content (#10141)

Browse source 
* build(model_prices_and_context_window.json): add vertex ai gemini-2.5-flash pricing

* build(model_prices_and_context_window.json): add gemini reasoning token pricing

* fix(vertex_and_google_ai_studio_gemini.py): support counting thinking tokens for gemini

allows accurate cost calc

* fix(utils.py): add reasoning token cost calc to generic cost calc

ensures gemini-2.5-flash cost calculation is accurate

* build(model_prices_and_context_window.json): mark gemini-2.5-flash as 'supports_reasoning'

* feat(gemini/): support 'thinking' + 'reasoning_effort' params + new unit tests

allow controlling thinking effort for gemini-2.5-flash models

* test: update unit testing

* feat(vertex_and_google_ai_studio_gemini.py): return reasoning content if given in gemini response

* test: update model name

* fix: fix ruff check

* test(test_spend_management_endpoints.py): update tests to be less sensitive to new keys / updates to usage object

* fix(vertex_and_google_ai_studio_gemini.py): fix translation
This commit is contained in:
Krish Dholakia 2025-04-19 09:20:52 -07:00 • committed by GitHub
parent db4ebe10c8
commit 36308a31be
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
16 changed files with 453 additions and 88 deletions
Download patch file Download diff file Expand all files Collapse all files

41
litellm/utils.py
View file

@ -516,9 +516,9 @@ def function_setup( # noqa: PLR0915
function_id: Optional[str] = kwargs["id"] if "id" in kwargs else None
## DYNAMIC CALLBACKS ##
dynamic_callbacks: Optional[List[Union[str, Callable, CustomLogger]]] = (
kwargs.pop("callbacks", None)
)
dynamic_callbacks: Optional[
List[Union[str, Callable, CustomLogger]]
] = kwargs.pop("callbacks", None)
all_callbacks = get_dynamic_callbacks(dynamic_callbacks=dynamic_callbacks)
if len(all_callbacks) > 0:
@ -1202,9 +1202,9 @@ def client(original_function): # noqa: PLR0915
exception=e,
retry_policy=kwargs.get("retry_policy"),
)
kwargs["retry_policy"] = (
reset_retry_policy()
) # prevent infinite loops
kwargs[
"retry_policy"
] = reset_retry_policy() # prevent infinite loops
litellm.num_retries = (
None # set retries to None to prevent infinite loops
)
@ -3013,16 +3013,16 @@ def get_optional_params( # noqa: PLR0915
True # so that main.py adds the function call to the prompt
)
if "tools" in non_default_params:
optional_params["functions_unsupported_model"] = (
non_default_params.pop("tools")
)
optional_params[
"functions_unsupported_model"
] = non_default_params.pop("tools")
non_default_params.pop(
"tool_choice", None
) # causes ollama requests to hang
elif "functions" in non_default_params:
optional_params["functions_unsupported_model"] = (
non_default_params.pop("functions")
)
optional_params[
"functions_unsupported_model"
] = non_default_params.pop("functions")
elif (
litellm.add_function_to_prompt
): # if user opts to add it to prompt instead
@ -3045,10 +3045,10 @@ def get_optional_params( # noqa: PLR0915
if "response_format" in non_default_params:
if provider_config is not None:
non_default_params["response_format"] = (
provider_config.get_json_schema_from_pydantic_object(
response_format=non_default_params["response_format"]
)
non_default_params[
"response_format"
] = provider_config.get_json_schema_from_pydantic_object(
response_format=non_default_params["response_format"]
)
else:
non_default_params["response_format"] = type_to_response_format_param(
@ -4064,9 +4064,9 @@ def _count_characters(text: str) -> int:
def get_response_string(response_obj: Union[ModelResponse, ModelResponseStream]) -> str:
_choices: Union[List[Union[Choices, StreamingChoices]], List[StreamingChoices]] = (
response_obj.choices
)
_choices: Union[
List[Union[Choices, StreamingChoices]], List[StreamingChoices]
] = response_obj.choices
response_str = ""
for choice in _choices:
@ -4563,6 +4563,9 @@ def _get_model_info_helper( # noqa: PLR0915
output_cost_per_character=_model_info.get(
"output_cost_per_character", None
),
output_cost_per_reasoning_token=_model_info.get(
"output_cost_per_reasoning_token", None
),
output_cost_per_token_above_128k_tokens=_model_info.get(
"output_cost_per_token_above_128k_tokens", None
),