litellm-mirror/litellm/llms/OpenAI/chat/o1_transformation.py
Krish Dholakia 234185ec13
LiteLLM Minor Fixes & Improvements (09/16/2024) (#5723) (#5731)
* LiteLLM Minor Fixes & Improvements (09/16/2024)  (#5723)

* coverage (#5713)

Signed-off-by: dbczumar <corey.zumar@databricks.com>

* Move (#5714)

Signed-off-by: dbczumar <corey.zumar@databricks.com>

* fix(litellm_logging.py): fix logging client re-init (#5710)

Fixes https://github.com/BerriAI/litellm/issues/5695

* fix(presidio.py): Fix logging_hook response and add support for additional presidio variables in guardrails config

Fixes https://github.com/BerriAI/litellm/issues/5682

* feat(o1_handler.py): fake streaming for openai o1 models

Fixes https://github.com/BerriAI/litellm/issues/5694

* docs: deprecated traceloop integration in favor of native otel (#5249)

* fix: fix linting errors

* fix: fix linting errors

* fix(main.py): fix o1 import

---------

Signed-off-by: dbczumar <corey.zumar@databricks.com>
Co-authored-by: Corey Zumar <39497902+dbczumar@users.noreply.github.com>
Co-authored-by: Nir Gazit <nirga@users.noreply.github.com>

* feat(spend_management_endpoints.py): expose `/global/spend/refresh` endpoint for updating material view (#5730)

* feat(spend_management_endpoints.py): expose `/global/spend/refresh` endpoint for updating material view

Supports having `MonthlyGlobalSpend` view be a material view, and exposes an endpoint to refresh it

* fix(custom_logger.py): reset calltype

* fix: fix linting errors

* fix: fix linting error

* fix: fix import

* test(test_databricks.py): fix databricks tests

---------

Signed-off-by: dbczumar <corey.zumar@databricks.com>
Co-authored-by: Corey Zumar <39497902+dbczumar@users.noreply.github.com>
Co-authored-by: Nir Gazit <nirga@users.noreply.github.com>
2024-09-17 08:05:52 -07:00

115 lines
3.8 KiB
Python

"""
Support for o1 model family
https://platform.openai.com/docs/guides/reasoning
Translations handled by LiteLLM:
- modalities: image => drop param (if user opts in to dropping param)
- role: system ==> translate to role 'user'
- streaming => faked by LiteLLM
- Tools, response_format => drop param (if user opts in to dropping param)
- Logprobs => drop param (if user opts in to dropping param)
"""
import types
from typing import Any, List, Optional, Union
import litellm
from litellm.types.llms.openai import AllMessageValues, ChatCompletionUserMessage
from .gpt_transformation import OpenAIGPTConfig
class OpenAIO1Config(OpenAIGPTConfig):
"""
Reference: https://platform.openai.com/docs/guides/reasoning
"""
@classmethod
def get_config(cls):
return {
k: v
for k, v in cls.__dict__.items()
if not k.startswith("__")
and not isinstance(
v,
(
types.FunctionType,
types.BuiltinFunctionType,
classmethod,
staticmethod,
),
)
and v is not None
}
def get_supported_openai_params(self, model: str) -> list:
"""
Get the supported OpenAI params for the given model
"""
all_openai_params = super().get_supported_openai_params(model=model)
non_supported_params = [
"logprobs",
"tools",
"tool_choice",
"parallel_tool_calls",
"function_call",
"functions",
"temperature",
"top_p",
"n",
"presence_penalty",
"frequency_penalty",
"top_logprobs",
"response_format",
"stop",
]
return [
param for param in all_openai_params if param not in non_supported_params
]
def map_openai_params(
self, non_default_params: dict, optional_params: dict, model: str
):
if "max_tokens" in non_default_params:
optional_params["max_completion_tokens"] = non_default_params.pop(
"max_tokens"
)
return super()._map_openai_params(non_default_params, optional_params, model)
def is_model_o1_reasoning_model(self, model: str) -> bool:
if model in litellm.open_ai_chat_completion_models and "o1" in model:
return True
return False
def o1_prompt_factory(self, messages: List[AllMessageValues]):
"""
Handles limitations of O-1 model family.
- modalities: image => drop param (if user opts in to dropping param)
- role: system ==> translate to role 'user'
"""
for i, message in enumerate(messages):
if message["role"] == "system":
new_message = ChatCompletionUserMessage(
content=message["content"], role="user"
)
messages[i] = new_message # Replace the old message with the new one
if "content" in message and isinstance(message["content"], list):
new_content = []
for content_item in message["content"]:
if content_item.get("type") == "image_url":
if litellm.drop_params is not True:
raise ValueError(
"Image content is not supported for O-1 models. Set litellm.drop_param to True to drop image content."
)
# If drop_param is True, we simply don't add the image content to new_content
else:
new_content.append(content_item)
message["content"] = new_content
return messages