phoenix/litellm-mirror
0
0
Fork 1
mirror of https://github.com/BerriAI/litellm.git synced 2025-04-26 03:04:13 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

LiteLLM Minor Fixes & Improvements (01/16/2025) - p2 (#7828)
All checks were successful
Read Version from pyproject.toml / read-version (push) Successful in 14s
Details

Browse source 
* fix(vertex_ai/gemini/transformation.py): handle 'http://' image urls

* test: add base test for `http:` url's

* fix(factory.py/get_image_details): follow redirects

allows http calls to work

* fix(codestral/): fix stream chunk parsing on last chunk of stream

* Azure ad token provider (#6917)

* Update azure.py

Added optional parameter azure ad token provider

* Added parameter to main.py

* Found token provider arg location

* Fixed embeddings

* Fixed ad token provider

---------

Co-authored-by: Krish Dholakia <krrishdholakia@gmail.com>

* fix: fix linting errors

* fix(main.py): leave out o1 route for azure ad token provider, for now

get v0 out for sync azure gpt route to begin with

* test: skip http:// test for fireworks ai

model does not support it

* refactor: cleanup dead code

* fix: revert http:// url passthrough for gemini

google ai studio raises errors

* test: fix test

---------

Co-authored-by: bahtman <anton@baht.dk>
This commit is contained in:
Krish Dholakia 2025-02-02 23:17:50 -08:00 • committed by GitHub
parent 10d3da7660
commit 97b8de17ab
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
9 changed files with 107 additions and 27 deletions
Download patch file Download diff file Expand all files Collapse all files

47
litellm/llms/azure/azure.py
View file

@ -2,7 +2,7 @@ import asyncio
import json
import os
import time
from typing import Any, Callable, List, Literal, Optional, Union
from typing import Any, Callable, Dict, List, Literal, Optional, Union
import httpx # type: ignore
from openai import AsyncAzureOpenAI, AzureOpenAI
@ -217,7 +217,7 @@ class AzureChatCompletion(BaseLLM):
def __init__(self) -> None:
super().__init__()
def validate_environment(self, api_key, azure_ad_token):
def validate_environment(self, api_key, azure_ad_token, azure_ad_token_provider):
headers = {
"content-type": "application/json",
}
@ -227,6 +227,10 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
headers["Authorization"] = f"Bearer {azure_ad_token}"
elif azure_ad_token_provider is not None:
azure_ad_token = azure_ad_token_provider()
headers["Authorization"] = f"Bearer {azure_ad_token}"
return headers
def _get_sync_azure_client(
@ -235,6 +239,7 @@ class AzureChatCompletion(BaseLLM):
api_base: Optional[str],
api_key: Optional[str],
azure_ad_token: Optional[str],
azure_ad_token_provider: Optional[Callable],
model: str,
max_retries: int,
timeout: Union[float, httpx.Timeout],
@ -242,7 +247,7 @@ class AzureChatCompletion(BaseLLM):
client_type: Literal["sync", "async"],
):
# init AzureOpenAI Client
azure_client_params = {
azure_client_params: Dict[str, Any] = {
"api_version": api_version,
"azure_endpoint": api_base,
"azure_deployment": model,
@ -259,6 +264,8 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = azure_ad_token_provider
if client is None:
if client_type == "sync":
azure_client = AzureOpenAI(**azure_client_params) # type: ignore
@ -326,6 +333,7 @@ class AzureChatCompletion(BaseLLM):
api_version: str,
api_type: str,
azure_ad_token: str,
azure_ad_token_provider: Callable,
dynamic_params: bool,
print_verbose: Callable,
timeout: Union[float, httpx.Timeout],
@ -373,6 +381,10 @@ class AzureChatCompletion(BaseLLM):
)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = (
azure_ad_token_provider
)
if acompletion is True:
client = AsyncAzureOpenAI(**azure_client_params)
@ -400,6 +412,7 @@ class AzureChatCompletion(BaseLLM):
api_key=api_key,
api_version=api_version,
azure_ad_token=azure_ad_token,
azure_ad_token_provider=azure_ad_token_provider,
timeout=timeout,
client=client,
)
@ -412,6 +425,7 @@ class AzureChatCompletion(BaseLLM):
api_version=api_version,
model=model,
azure_ad_token=azure_ad_token,
azure_ad_token_provider=azure_ad_token_provider,
dynamic_params=dynamic_params,
timeout=timeout,
client=client,
@ -428,6 +442,7 @@ class AzureChatCompletion(BaseLLM):
api_key=api_key,
api_version=api_version,
azure_ad_token=azure_ad_token,
azure_ad_token_provider=azure_ad_token_provider,
timeout=timeout,
client=client,
)
@ -468,6 +483,10 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = (
azure_ad_token_provider
)
if (
client is None
@ -535,6 +554,7 @@ class AzureChatCompletion(BaseLLM):
model_response: ModelResponse,
logging_obj: LiteLLMLoggingObj,
azure_ad_token: Optional[str] = None,
azure_ad_token_provider: Optional[Callable] = None,
convert_tool_call_to_json_mode: Optional[bool] = None,
client=None, # this is the AsyncAzureOpenAI
):
@ -564,6 +584,8 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = azure_ad_token_provider
# setting Azure client
if client is None or dynamic_params:
@ -650,6 +672,7 @@ class AzureChatCompletion(BaseLLM):
model: str,
timeout: Any,
azure_ad_token: Optional[str] = None,
azure_ad_token_provider: Optional[Callable] = None,
client=None,
):
max_retries = data.pop("max_retries", 2)
@ -675,6 +698,8 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = azure_ad_token_provider
if client is None or dynamic_params:
azure_client = AzureOpenAI(**azure_client_params)
@ -718,6 +743,7 @@ class AzureChatCompletion(BaseLLM):
model: str,
timeout: Any,
azure_ad_token: Optional[str] = None,
azure_ad_token_provider: Optional[Callable] = None,
client=None,
):
try:
@ -739,6 +765,8 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = azure_ad_token_provider
if client is None or dynamic_params:
azure_client = AsyncAzureOpenAI(**azure_client_params)
else:
@ -844,6 +872,7 @@ class AzureChatCompletion(BaseLLM):
optional_params: dict,
api_key: Optional[str] = None,
azure_ad_token: Optional[str] = None,
azure_ad_token_provider: Optional[Callable] = None,
max_retries: Optional[int] = None,
client=None,
aembedding=None,
@ -883,6 +912,8 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = azure_ad_token_provider
## LOGGING
logging_obj.pre_call(
@ -1240,6 +1271,7 @@ class AzureChatCompletion(BaseLLM):
api_version: Optional[str] = None,
model_response: Optional[ImageResponse] = None,
azure_ad_token: Optional[str] = None,
azure_ad_token_provider: Optional[Callable] = None,
client=None,
aimg_generation=None,
) -> ImageResponse:
@ -1266,7 +1298,7 @@ class AzureChatCompletion(BaseLLM):
)
# init AzureOpenAI Client
azure_client_params = {
azure_client_params: Dict[str, Any] = {
"api_version": api_version,
"azure_endpoint": api_base,
"azure_deployment": model,
@ -1282,6 +1314,8 @@ class AzureChatCompletion(BaseLLM):
if azure_ad_token.startswith("oidc/"):
azure_ad_token = get_azure_ad_token_from_oidc(azure_ad_token)
azure_client_params["azure_ad_token"] = azure_ad_token
elif azure_ad_token_provider is not None:
azure_client_params["azure_ad_token_provider"] = azure_ad_token_provider
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) # type: ignore
@ -1342,6 +1376,7 @@ class AzureChatCompletion(BaseLLM):
max_retries: int,
timeout: Union[float, httpx.Timeout],
azure_ad_token: Optional[str] = None,
azure_ad_token_provider: Optional[Callable] = None,
aspeech: Optional[bool] = None,
client=None,
) -> HttpxBinaryResponseContent:
@ -1358,6 +1393,7 @@ class AzureChatCompletion(BaseLLM):
api_base=api_base,
api_version=api_version,
azure_ad_token=azure_ad_token,
azure_ad_token_provider=azure_ad_token_provider,
max_retries=max_retries,
timeout=timeout,
client=client,
@ -1368,6 +1404,7 @@ class AzureChatCompletion(BaseLLM):
api_version=api_version,
api_key=api_key,
azure_ad_token=azure_ad_token,
azure_ad_token_provider=azure_ad_token_provider,
model=model,
max_retries=max_retries,
timeout=timeout,
@ -1393,6 +1430,7 @@ class AzureChatCompletion(BaseLLM):
api_base: Optional[str],
api_version: Optional[str],
azure_ad_token: Optional[str],
azure_ad_token_provider: Optional[Callable],
max_retries: int,
timeout: Union[float, httpx.Timeout],
client=None,
@ -1403,6 +1441,7 @@ class AzureChatCompletion(BaseLLM):
api_version=api_version,
api_key=api_key,
azure_ad_token=azure_ad_token,
azure_ad_token_provider=azure_ad_token_provider,
model=model,
max_retries=max_retries,
timeout=timeout,