forked from phoenix/litellm-mirror
fix - refactor proxy server to use common func
This commit is contained in:
Ishaan Jaff
parent
58eb352ddb
commit
b5162438c6
2 changed files with 144 additions and 696 deletions
|
|
@ -4146,6 +4146,7 @@ async def chat_completion(
|
|||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
|
||||
data["model"] = (
|
||||
|
|
@ -4410,7 +4411,6 @@ async def completion(
|
|||
except:
|
||||
data = json.loads(body_str)
|
||||
|
||||
data["user"] = data.get("user", user_api_key_dict.user_id)
|
||||
data["model"] = (
|
||||
general_settings.get("completion_model", None) # server default
|
||||
or user_model # model name passed via cli args
|
||||
|
|
@ -4419,30 +4419,15 @@ async def completion(
|
|||
)
|
||||
if user_model:
|
||||
data["model"] = user_model
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
# override with user settings, these are params passed via cli
|
||||
if user_temperature:
|
||||
|
|
@ -4642,15 +4627,14 @@ async def embeddings(
|
|||
data = orjson.loads(body)
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = {
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
|
||||
data["model"] = (
|
||||
general_settings.get("embedding_model", None) # server default
|
||||
|
|
@ -4660,45 +4644,6 @@ async def embeddings(
|
|||
)
|
||||
if user_model:
|
||||
data["model"] = user_model
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
### MODEL ALIAS MAPPING ###
|
||||
# check if model name in model alias map
|
||||
|
|
@ -4853,15 +4798,14 @@ async def image_generation(
|
|||
data = orjson.loads(body)
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = {
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
|
||||
data["model"] = (
|
||||
general_settings.get("image_generation_model", None) # server default
|
||||
|
|
@ -4871,46 +4815,6 @@ async def image_generation(
|
|||
if user_model:
|
||||
data["model"] = user_model
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
### MODEL ALIAS MAPPING ###
|
||||
# check if model name in model alias map
|
||||
# get the actual model name
|
||||
|
|
@ -5035,12 +4939,14 @@ async def audio_speech(
|
|||
data = orjson.loads(body)
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
|
@ -5048,46 +4954,6 @@ async def audio_speech(
|
|||
if user_model:
|
||||
data["model"] = user_model
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
router_model_names = llm_router.model_names if llm_router is not None else []
|
||||
|
||||
### CALL HOOKS ### - modify incoming data / reject request before calling the model
|
||||
|
|
@ -5200,12 +5066,14 @@ async def audio_transcriptions(
|
|||
data = {key: value for key, value in form_data.items() if key != "file"}
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
|
@ -5218,47 +5086,6 @@ async def audio_transcriptions(
|
|||
if user_model:
|
||||
data["model"] = user_model
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
data["metadata"]["file_name"] = file.filename
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
router_model_names = llm_router.model_names if llm_router is not None else []
|
||||
|
||||
assert (
|
||||
|
|
@ -5410,55 +5237,14 @@ async def get_assistants(
|
|||
body = await request.body()
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
# for now use custom_llm_provider=="openai" -> this will change as LiteLLM adds more providers for acreate_batch
|
||||
if llm_router is None:
|
||||
|
|
@ -5539,55 +5325,14 @@ async def create_threads(
|
|||
body = await request.body()
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "litellm_metadata" not in data:
|
||||
data["litellm_metadata"] = {}
|
||||
data["litellm_metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["litellm_metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["litellm_metadata"]["headers"] = _headers
|
||||
data["litellm_metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["litellm_metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["litellm_metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["litellm_metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["litellm_metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["litellm_metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["litellm_metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
# for now use custom_llm_provider=="openai" -> this will change as LiteLLM adds more providers for acreate_batch
|
||||
if llm_router is None:
|
||||
|
|
@ -5667,55 +5412,14 @@ async def get_thread(
|
|||
try:
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
# for now use custom_llm_provider=="openai" -> this will change as LiteLLM adds more providers for acreate_batch
|
||||
if llm_router is None:
|
||||
|
|
@ -5798,55 +5502,14 @@ async def add_messages(
|
|||
data = orjson.loads(body)
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "litellm_metadata" not in data:
|
||||
data["litellm_metadata"] = {}
|
||||
data["litellm_metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["litellm_metadata"]["litellm_api_version"] = version
|
||||
data["litellm_metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["litellm_metadata"]["headers"] = _headers
|
||||
data["litellm_metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["litellm_metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["litellm_metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["litellm_metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["litellm_metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["litellm_metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["litellm_metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
# for now use custom_llm_provider=="openai" -> this will change as LiteLLM adds more providers for acreate_batch
|
||||
if llm_router is None:
|
||||
|
|
@ -5925,55 +5588,14 @@ async def get_messages(
|
|||
data: Dict = {}
|
||||
try:
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
# for now use custom_llm_provider=="openai" -> this will change as LiteLLM adds more providers for acreate_batch
|
||||
if llm_router is None:
|
||||
|
|
@ -6054,55 +5676,14 @@ async def run_thread(
|
|||
body = await request.body()
|
||||
data = orjson.loads(body)
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "litellm_metadata" not in data:
|
||||
data["litellm_metadata"] = {}
|
||||
data["litellm_metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["litellm_metadata"]["litellm_api_version"] = version
|
||||
data["litellm_metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["litellm_metadata"]["headers"] = _headers
|
||||
data["litellm_metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["litellm_metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["litellm_metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["litellm_metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["litellm_metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["litellm_metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["litellm_metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
# for now use custom_llm_provider=="openai" -> this will change as LiteLLM adds more providers for acreate_batch
|
||||
if llm_router is None:
|
||||
|
|
@ -6214,55 +5795,14 @@ async def create_batch(
|
|||
data = {key: value for key, value in form_data.items() if key != "file"}
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
_create_batch_data = CreateBatchRequest(**data)
|
||||
|
||||
|
|
@ -6355,55 +5895,14 @@ async def retrieve_batch(
|
|||
data = {key: value for key, value in form_data.items() if key != "file"}
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
_retrieve_batch_request = RetrieveBatchRequest(
|
||||
batch_id=batch_id,
|
||||
|
|
@ -6510,55 +6009,14 @@ async def create_file(
|
|||
data = {key: value for key, value in form_data.items() if key != "file"}
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = { # type: ignore
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
_create_file_request = CreateFileRequest()
|
||||
|
||||
|
|
@ -6648,15 +6106,14 @@ async def moderations(
|
|||
data = orjson.loads(body)
|
||||
|
||||
# Include original request and headers in the data
|
||||
data["proxy_server_request"] = {
|
||||
"url": str(request.url),
|
||||
"method": request.method,
|
||||
"headers": dict(request.headers),
|
||||
"body": copy.copy(data), # use copy instead of deepcopy
|
||||
}
|
||||
|
||||
if data.get("user", None) is None and user_api_key_dict.user_id is not None:
|
||||
data["user"] = user_api_key_dict.user_id
|
||||
data = await add_litellm_data_to_request(
|
||||
data=data,
|
||||
request=request,
|
||||
general_settings=general_settings,
|
||||
user_api_key_dict=user_api_key_dict,
|
||||
version=version,
|
||||
proxy_config=proxy_config,
|
||||
)
|
||||
|
||||
data["model"] = (
|
||||
general_settings.get("moderation_model", None) # server default
|
||||
|
|
@ -6666,46 +6123,6 @@ async def moderations(
|
|||
if user_model:
|
||||
data["model"] = user_model
|
||||
|
||||
if "metadata" not in data:
|
||||
data["metadata"] = {}
|
||||
data["metadata"]["user_api_key"] = user_api_key_dict.api_key
|
||||
data["metadata"]["litellm_api_version"] = version
|
||||
data["metadata"]["user_api_key_metadata"] = user_api_key_dict.metadata
|
||||
_headers = dict(request.headers)
|
||||
_headers.pop(
|
||||
"authorization", None
|
||||
) # do not store the original `sk-..` api key in the db
|
||||
data["metadata"]["headers"] = _headers
|
||||
data["metadata"]["global_max_parallel_requests"] = general_settings.get(
|
||||
"global_max_parallel_requests", None
|
||||
)
|
||||
data["metadata"]["user_api_key_alias"] = getattr(
|
||||
user_api_key_dict, "key_alias", None
|
||||
)
|
||||
data["metadata"]["user_api_key_user_id"] = user_api_key_dict.user_id
|
||||
data["metadata"]["user_api_key_team_id"] = getattr(
|
||||
user_api_key_dict, "team_id", None
|
||||
)
|
||||
data["metadata"]["user_api_key_team_alias"] = getattr(
|
||||
user_api_key_dict, "team_alias", None
|
||||
)
|
||||
data["metadata"]["endpoint"] = str(request.url)
|
||||
|
||||
### TEAM-SPECIFIC PARAMS ###
|
||||
if user_api_key_dict.team_id is not None:
|
||||
team_config = await proxy_config.load_team_config(
|
||||
team_id=user_api_key_dict.team_id
|
||||
)
|
||||
if len(team_config) == 0:
|
||||
pass
|
||||
else:
|
||||
team_id = team_config.pop("team_id", None)
|
||||
data["metadata"]["team_id"] = team_id
|
||||
data = {
|
||||
**team_config,
|
||||
**data,
|
||||
} # add the team-specific configs to the completion call
|
||||
|
||||
router_model_names = llm_router.model_names if llm_router is not None else []
|
||||
|
||||
### CALL HOOKS ### - modify incoming data / reject request before calling the model
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue