fix imports

litellm/integrations/SlackAlerting/slack_alerting.py

@@ -27,7 +27,7 @@ from litellm.llms.custom_httpx.http_handler import (
     httpxSpecialProvider,
 )
 from litellm.types.integrations.slack_alerting import *
-from litellm_proxy._types import AlertType, CallInfo, VirtualKeyEvent, WebhookEvent
+from litellm.types.utils import CallInfo, VirtualKeyEvent, WebhookEvent
 
 from ..email_templates.templates import *
 from .batching_handler import send_to_webhook, squash_payloads

litellm/integrations/SlackAlerting/utils.py

@@ -6,7 +6,7 @@ import asyncio
 from typing import TYPE_CHECKING, Any, Dict, List, Optional, Union
 
 from litellm.secret_managers.main import get_secret
-from litellm_proxy._types import AlertType
+from litellm.types.integrations.slack_alerting import AlertType
 
 if TYPE_CHECKING:
     from litellm.litellm_core_utils.litellm_logging import Logging as _Logging

litellm/integrations/gcs_bucket/gcs_bucket.py

@@ -12,7 +12,6 @@ from litellm.integrations.gcs_bucket.gcs_bucket_base import GCSBucketBase
 from litellm.types.integrations.base_health_check import IntegrationHealthCheckStatus
 from litellm.types.integrations.gcs_bucket import *
 from litellm.types.utils import StandardLoggingPayload
-from litellm_proxy._types import CommonProxyErrors
 
 if TYPE_CHECKING:
     from litellm.llms.vertex_ai.vertex_llm_base import VertexBase
@@ -22,7 +21,7 @@ else:
 
 class GCSBucketLogger(GCSBucketBase, AdditionalLoggingUtils):
     def __init__(self, bucket_name: Optional[str] = None) -> None:
-        from litellm_proxy.proxy_server import premium_user
+        from litellm_proxy.proxy_server import CommonProxyErrors, premium_user
 
         super().__init__(bucket_name=bucket_name)
 
@@ -48,7 +47,7 @@ class GCSBucketLogger(GCSBucketBase, AdditionalLoggingUtils):
 
     #### ASYNC ####
     async def async_log_success_event(self, kwargs, response_obj, start_time, end_time):
-        from litellm_proxy.proxy_server import premium_user
+        from litellm_proxy.proxy_server import CommonProxyErrors, premium_user
 
         if premium_user is not True:
             raise ValueError(

litellm/integrations/prometheus.py

@@ -21,12 +21,15 @@ from litellm.integrations.custom_logger import CustomLogger
 from litellm.types.integrations.prometheus import *
 from litellm.types.utils import StandardLoggingPayload
 from litellm.utils import get_end_user_id_for_cost_tracking
-from litellm_proxy._types import LiteLLM_TeamTable, UserAPIKeyAuth
 
 if TYPE_CHECKING:
     from apscheduler.schedulers.asyncio import AsyncIOScheduler
+
+    from litellm_proxy._types import LiteLLM_TeamTable, UserAPIKeyAuth
 else:
     AsyncIOScheduler = Any
+    UserAPIKeyAuth = Any
+    LiteLLM_TeamTable = Any
 
 
 class PrometheusLogger(CustomLogger):
@@ -1511,6 +1514,8 @@ class PrometheusLogger(CustomLogger):
         self, keys: List[Union[str, UserAPIKeyAuth]]
     ):
         """Helper function to set budget metrics for a list of keys"""
+        from litellm_proxy._types import UserAPIKeyAuth
+
         for key in keys:
             if isinstance(key, UserAPIKeyAuth):
                 self._set_key_budget_metrics(key)

litellm/secret_managers/aws_secret_manager.py

@@ -15,7 +15,7 @@ import re
 from typing import Any, Dict, Optional
 
 import litellm
-from litellm_proxy._types import KeyManagementSystem
+from litellm.types.utils import KeyManagementSystem
 
 
 def validate_environment():

litellm/secret_managers/aws_secret_manager_v2.py

@@ -27,7 +27,7 @@ from litellm.llms.custom_httpx.http_handler import (
     get_async_httpx_client,
 )
 from litellm.types.llms.custom_http import httpxSpecialProvider
-from litellm_proxy._types import KeyManagementSystem
+from litellm.types.utils import KeyManagementSystem
 
 from .base_secret_manager import BaseSecretManager
 

litellm/secret_managers/google_kms.py

@@ -12,7 +12,7 @@ import os
 from typing import Optional
 
 import litellm
-from litellm_proxy._types import KeyManagementSystem
+from litellm.types.utils import KeyManagementSystem
 
 
 def validate_environment():

litellm/secret_managers/google_secret_manager.py

@@ -8,7 +8,7 @@ from litellm.caching.caching import InMemoryCache
 from litellm.constants import SECRET_MANAGER_REFRESH_INTERVAL
 from litellm.integrations.gcs_bucket.gcs_bucket_base import GCSBucketBase
 from litellm.llms.custom_httpx.http_handler import _get_httpx_client
-from litellm_proxy._types import CommonProxyErrors, KeyManagementSystem
+from litellm.types.utils import KeyManagementSystem
 
 
 class GoogleSecretManager(GCSBucketBase):
@@ -22,7 +22,7 @@ class GoogleSecretManager(GCSBucketBase):
             refresh_interval (int, optional): The refresh interval in seconds. Defaults to 86400. (24 hours)
             always_read_secret_manager (bool, optional): Whether to always read from the secret manager. Defaults to False. Since we do want to cache values
         """
-        from litellm_proxy.proxy_server import premium_user
+        from litellm_proxy.proxy_server import CommonProxyErrors, premium_user
 
         if premium_user is not True:
             raise ValueError(

litellm/secret_managers/hashicorp_secret_manager.py

@@ -12,7 +12,7 @@ from litellm.llms.custom_httpx.http_handler import (
     get_async_httpx_client,
     httpxSpecialProvider,
 )
-from litellm_proxy._types import KeyManagementSystem
+from litellm.types.utils import KeyManagementSystem
 
 from .base_secret_manager import BaseSecretManager
 

litellm/types/proxy/management_endpoints/internal_user_endpoints.py

@@ -1,9 +1,12 @@
-from typing import Any, Dict, List, Literal, Optional, Union
+from typing import TYPE_CHECKING, Any, Dict, List, Literal, Optional, Union
 
 from fastapi import HTTPException
 from pydantic import BaseModel, EmailStr
 
-from litellm_proxy._types import LiteLLM_UserTableWithKeyCount
+if TYPE_CHECKING:
+    from litellm_proxy._types import LiteLLM_UserTableWithKeyCount
+else:
+    LiteLLM_UserTableWithKeyCount = Any
 
 
 class UserListResponse(BaseModel):

litellm/types/utils.py

@@ -2321,3 +2321,41 @@ class LiteLLM_UpperboundKeyGenerateParams(LiteLLMPydanticObjectBase):
     max_parallel_requests: Optional[int] = None
     tpm_limit: Optional[int] = None
     rpm_limit: Optional[int] = None
+
+
+class CallInfo(LiteLLMPydanticObjectBase):
+    """Used for slack budget alerting"""
+
+    spend: float
+    max_budget: Optional[float] = None
+    soft_budget: Optional[float] = None
+    token: Optional[str] = Field(default=None, description="Hashed value of that key")
+    customer_id: Optional[str] = None
+    user_id: Optional[str] = None
+    team_id: Optional[str] = None
+    team_alias: Optional[str] = None
+    user_email: Optional[str] = None
+    key_alias: Optional[str] = None
+    projected_exceeded_date: Optional[str] = None
+    projected_spend: Optional[float] = None
+
+
+class WebhookEvent(CallInfo):
+    event: Literal[
+        "budget_crossed",
+        "soft_budget_crossed",
+        "threshold_crossed",
+        "projected_limit_exceeded",
+        "key_created",
+        "internal_user_created",
+        "spend_tracked",
+    ]
+    event_group: Literal["internal_user", "key", "team", "proxy", "customer"]
+    event_message: str  # human-readable description of event
+
+
+class VirtualKeyEvent(LiteLLMPydanticObjectBase):
+    created_by_user_id: str
+    created_by_user_role: str
+    created_by_key_alias: Optional[str]
+    request_kwargs: dict

litellm_proxy/_types.py

@@ -38,6 +38,7 @@ from litellm.types.utils import (
     KeyManagementSettings, # noqa: F401
     LiteLLM_UpperboundKeyGenerateParams # noqa: F401
 )
+from litellm.types.utils import CallInfo, WebhookEvent, VirtualKeyEvent # noqa: F401
 
 from .types_utils.utils import get_instance_fn, validate_custom_validate_return_type
 
@@ -1745,37 +1746,6 @@ class TokenCountResponse(LiteLLMPydanticObjectBase):
     tokenizer_type: str
 
 
-class CallInfo(LiteLLMPydanticObjectBase):
-    """Used for slack budget alerting"""
-
-    spend: float
-    max_budget: Optional[float] = None
-    soft_budget: Optional[float] = None
-    token: Optional[str] = Field(default=None, description="Hashed value of that key")
-    customer_id: Optional[str] = None
-    user_id: Optional[str] = None
-    team_id: Optional[str] = None
-    team_alias: Optional[str] = None
-    user_email: Optional[str] = None
-    key_alias: Optional[str] = None
-    projected_exceeded_date: Optional[str] = None
-    projected_spend: Optional[float] = None
-
-
-class WebhookEvent(CallInfo):
-    event: Literal[
-        "budget_crossed",
-        "soft_budget_crossed",
-        "threshold_crossed",
-        "projected_limit_exceeded",
-        "key_created",
-        "internal_user_created",
-        "spend_tracked",
-    ]
-    event_group: Literal["internal_user", "key", "team", "proxy", "customer"]
-    event_message: str  # human-readable description of event
-
-
 class SpecialModelNames(enum.Enum):
     all_team_models = "all-team-models"
     all_proxy_models = "all-proxy-models"
@@ -2134,13 +2104,6 @@ class SSOUserDefinedValues(TypedDict):
     budget_duration: Optional[str]
 
 
-class VirtualKeyEvent(LiteLLMPydanticObjectBase):
-    created_by_user_id: str
-    created_by_user_role: str
-    created_by_key_alias: Optional[str]
-    request_kwargs: dict
-
-
 class CreatePassThroughEndpoint(LiteLLMPydanticObjectBase):
     path: str
     target: str