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

LiteLLM Minor Fixes & Improvements (10/07/2024) (#6101)

Browse source 
* fix(utils.py): support dropping temperature param for azure o1 models

* fix(main.py): handle azure o1 streaming requests

o1 doesn't support streaming, fake it to ensure code works as expected

* feat(utils.py): expose `hosted_vllm/` endpoint, with tool handling for vllm

Fixes https://github.com/BerriAI/litellm/issues/6088

* refactor(internal_user_endpoints.py): cleanup unused params + update docstring

Closes https://github.com/BerriAI/litellm/issues/6100

* fix(main.py): expose custom image generation api support

Fixes https://github.com/BerriAI/litellm/issues/6097

* fix: fix linting errors

* docs(custom_llm_server.md): add docs on custom api for image gen calls

* fix(types/utils.py): handle dict type

* fix(types/utils.py): fix linting errors
This commit is contained in:
Krish Dholakia 2024-10-08 01:17:22 -04:00 • committed by GitHub
parent 5de69cb1b2
commit 6729c9ca7f
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
17 changed files with 643 additions and 76 deletions
Download patch file Download diff file Expand all files Collapse all files

96
litellm/utils.py
View file

@ -2788,6 +2788,24 @@ def _remove_additional_properties(schema):
return schema
def _remove_strict_from_schema(schema):
if isinstance(schema, dict):
# Remove the 'additionalProperties' key if it exists and is set to False
if "strict" in schema:
del schema["strict"]
# Recursively process all dictionary values
for key, value in schema.items():
_remove_strict_from_schema(value)
elif isinstance(schema, list):
# Recursively process all items in the list
for item in schema:
_remove_strict_from_schema(item)
return schema
def get_optional_params(
# use the openai defaults
# https://platform.openai.com/docs/api-reference/chat/create
@ -2999,13 +3017,19 @@ def get_optional_params(
from litellm.llms.vertex_ai_and_google_ai_studio.common_utils import (
_build_vertex_schema,
)
old_schema = copy.deepcopy(
non_default_params["response_format"]
.get("json_schema", {})
.get("schema")
)
new_schema = _remove_additional_properties(schema=old_schema)
new_schema = _build_vertex_schema(parameters=new_schema)
if isinstance(new_schema, list):
for item in new_schema:
if isinstance(item, dict):
item = _build_vertex_schema(parameters=item)
elif isinstance(new_schema, dict):
new_schema = _build_vertex_schema(parameters=new_schema)
non_default_params["response_format"]["json_schema"]["schema"] = new_schema
if "tools" in non_default_params and isinstance(
non_default_params, list
@ -3767,6 +3791,21 @@ def get_optional_params(
optional_params=optional_params,
model=model,
)
elif custom_llm_provider == "hosted_vllm":
supported_params = get_supported_openai_params(
model=model, custom_llm_provider=custom_llm_provider
)
_check_valid_arg(supported_params=supported_params)
optional_params = litellm.HostedVLLMChatConfig().map_openai_params(
non_default_params=non_default_params,
optional_params=optional_params,
model=model,
drop_params=(
drop_params
if drop_params is not None and isinstance(drop_params, bool)
else False
),
)
elif custom_llm_provider == "groq":
supported_params = get_supported_openai_params(
@ -3926,24 +3965,36 @@ def get_optional_params(
model=model, custom_llm_provider="azure"
)
_check_valid_arg(supported_params=supported_params)
verbose_logger.debug(
"Azure optional params - api_version: api_version={}, litellm.api_version={}, os.environ['AZURE_API_VERSION']={}".format(
api_version, litellm.api_version, get_secret("AZURE_API_VERSION")
if litellm.AzureOpenAIO1Config().is_o1_model(model=model):
optional_params = litellm.AzureOpenAIO1Config().map_openai_params(
non_default_params=non_default_params,
optional_params=optional_params,
model=model,
drop_params=(
drop_params
if drop_params is not None and isinstance(drop_params, bool)
else False
),
)
else:
verbose_logger.debug(
"Azure optional params - api_version: api_version={}, litellm.api_version={}, os.environ['AZURE_API_VERSION']={}".format(
api_version, litellm.api_version, get_secret("AZURE_API_VERSION")
)
)
api_version = (
api_version
or litellm.api_version
or get_secret("AZURE_API_VERSION")
or litellm.AZURE_DEFAULT_API_VERSION
)
optional_params = litellm.AzureOpenAIConfig().map_openai_params(
non_default_params=non_default_params,
optional_params=optional_params,
model=model,
api_version=api_version, # type: ignore
drop_params=drop_params,
)
)
api_version = (
api_version
or litellm.api_version
or get_secret("AZURE_API_VERSION")
or litellm.AZURE_DEFAULT_API_VERSION
)
optional_params = litellm.AzureOpenAIConfig().map_openai_params(
non_default_params=non_default_params,
optional_params=optional_params,
model=model,
api_version=api_version, # type: ignore
drop_params=drop_params,
)
else: # assume passing in params for text-completion openai
supported_params = get_supported_openai_params(
model=model, custom_llm_provider="custom_openai"
@ -4409,6 +4460,8 @@ def get_supported_openai_params(
"extra_headers",
"extra_body",
]
elif custom_llm_provider == "hosted_vllm":
return litellm.HostedVLLMChatConfig().get_supported_openai_params(model=model)
elif custom_llm_provider == "deepseek":
return [
# https://platform.deepseek.com/api-docs/api/create-chat-completion
@ -4465,7 +4518,12 @@ def get_supported_openai_params(
elif custom_llm_provider == "openai":
return litellm.OpenAIConfig().get_supported_openai_params(model=model)
elif custom_llm_provider == "azure":
return litellm.AzureOpenAIConfig().get_supported_openai_params()
if litellm.AzureOpenAIO1Config().is_o1_model(model=model):
return litellm.AzureOpenAIO1Config().get_supported_openai_params(
model=model
)
else:
return litellm.AzureOpenAIConfig().get_supported_openai_params()
elif custom_llm_provider == "openrouter":
return [
"temperature",