phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-25 10:44:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Complete o3 model support (#8183)

Browse source 
* fix(o_series_transformation.py): add 'reasoning_effort' as o series model param

Closes https://github.com/BerriAI/litellm/issues/8182

* fix(main.py): ensure `reasoning_effort` is a mapped openai param

* refactor(azure/): rename o1_[x] files to o_series_[x]

* refactor(base_llm_unit_tests.py): refactor testing for o series reasoning effort

* test(test_azure_o_series.py): have azure o series tests correctly inherit from base o series model tests

* feat(base_utils.py): support translating 'developer' role to 'system' role for non-openai providers

Makes it easy to switch from openai to anthropic

* fix: fix linting errors

* fix(base_llm_unit_tests.py): fix test

* fix(main.py): add missing param
This commit is contained in:
Krish Dholakia 2025-02-02 22:36:37 -08:00 • committed by GitHub
parent e4566d7b1c
commit 1105e35538
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
15 changed files with 230 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

20
litellm/main.py
View file

@ -67,6 +67,7 @@ from litellm.litellm_core_utils.mock_functions import (
from litellm.litellm_core_utils.prompt_templates.common_utils import (
get_content_from_model_response,
)
from litellm.llms.base_llm.chat.transformation import BaseConfig
from litellm.llms.custom_httpx.http_handler import AsyncHTTPHandler, HTTPHandler
from litellm.realtime_api.main import _realtime_health_check
from litellm.secret_managers.main import get_secret_str
@ -115,7 +116,7 @@ from .llms import baseten, maritalk, ollama_chat
from .llms.anthropic.chat import AnthropicChatCompletion
from .llms.azure.audio_transcriptions import AzureAudioTranscription
from .llms.azure.azure import AzureChatCompletion, _check_dynamic_azure_params
from .llms.azure.chat.o1_handler import AzureOpenAIO1ChatCompletion
from .llms.azure.chat.o_series_handler import AzureOpenAIO1ChatCompletion
from .llms.azure.completion.handler import AzureTextCompletion
from .llms.azure_ai.embed import AzureAIEmbedding
from .llms.bedrock.chat import BedrockConverseLLM, BedrockLLM
@ -331,6 +332,7 @@ async def acompletion(
logprobs: Optional[bool] = None,
top_logprobs: Optional[int] = None,
deployment_id=None,
reasoning_effort: Optional[Literal["low", "medium", "high"]] = None,
# set api_base, api_version, api_key
base_url: Optional[str] = None,
api_version: Optional[str] = None,
@ -425,6 +427,7 @@ async def acompletion(
"api_version": api_version,
"api_key": api_key,
"model_list": model_list,
"reasoning_effort": reasoning_effort,
"extra_headers": extra_headers,
"acompletion": True, # assuming this is a required parameter
}
@ -777,6 +780,7 @@ def completion( # type: ignore # noqa: PLR0915
logit_bias: Optional[dict] = None,
user: Optional[str] = None,
# openai v1.0+ new params
reasoning_effort: Optional[Literal["low", "medium", "high"]] = None,
response_format: Optional[Union[dict, Type[BaseModel]]] = None,
seed: Optional[int] = None,
tools: Optional[List] = None,
@ -1046,6 +1050,19 @@ def completion( # type: ignore # noqa: PLR0915
if eos_token:
custom_prompt_dict[model]["eos_token"] = eos_token
provider_config: Optional[BaseConfig] = None
if custom_llm_provider is not None and custom_llm_provider in [
provider.value for provider in LlmProviders
]:
provider_config = ProviderConfigManager.get_provider_chat_config(
model=model, provider=LlmProviders(custom_llm_provider)
)
if provider_config is not None:
messages = provider_config.translate_developer_role_to_system_role(
messages=messages
)
if (
supports_system_message is not None
and isinstance(supports_system_message, bool)
@ -1087,6 +1104,7 @@ def completion( # type: ignore # noqa: PLR0915
api_version=api_version,
parallel_tool_calls=parallel_tool_calls,
messages=messages,
reasoning_effort=reasoning_effort,
**non_default_params,
)