Query the DB for managed IDs the caller owns and return an OpenAI-style paginated list response. Returns ``None`` when ``prisma_client`` is u

Get a paginated list of teams with filtering and sorting options. Parameters: user_id: Optional[str] Only return teams w

Performs the actual redaction on the logging object and result.

Main entry point - find and rotate keys that are due for rotation. Uses PodLockManager to ensure only one pod runs rotation in multi-

Process SSO JWT access token and extract team IDs and user role if available. This function decodes the JWT access token and extracts team I

Render a prompt template with the given variables. Args: prompt_id: The ID of the prompt template to render

Resolve the final guardrails for a single policy, including inheritance. This method: 1. Resolves the inheritance chain

Recursively walk a request body dict/list and resolve any passthrough managed IDs. Skips litellm internal keys (``litellm_*``). Returns

Safely sets an attribute on the span, ensuring the value is a primitive type.

Sanitize messages for tool calling to handle common issues when modify_params=True: Case A: Missing tool_result for tool_use (orphaned tool

Receive message from backend and store it

Converts the ProxyException instance to a dictionary.

Args: request: TokenCountRequest call_endpoint: bool - When set to "True" it will call the token counting endpoint - e.g Anthropi

Transform a Chat Completion response into a Responses API response

Transform OpenAI embedding input to Gemini batchEmbedContents format. Each input element becomes a separate EmbedContentRequest, supporting

Transform OpenAI embedding input to Gemini embedContent format (multimodal). Args: input: GeminiEmbeddingInput with text, data URIs,

Transform request to provider-specific format. Returns: TextToSpeechRequestData: A structured dict containing:

Transform the Veo video creation response. Veo returns: { "name": "operations/generate_1234567890",

Update the polling state in Redis Uses OpenAI Response object format with native status types: https://platform.openai.com/d

Add proxy hooks to litellm.callbacks

Call a regular MCP tool using the MCP client. Args: mcp_server: The MCP server configuration original_tool_n

Run all cases, print confusion matrix, save results JSON.

Helper to get AsyncOpenAI or AsyncAzureOpenAI client that was created for the deployment The same OpenAI client is re-used to optimi

Compute CORS allowed origins and credentials flag from environment variables. Extracted into a function so it can be unit-tested without rel

Extract a semantic-cache prompt from chat or Responses API request kwargs.

Get SSL verification setting for boto3 clients. This ensures that custom CA certificates are properly used for all AWS API calls,

Return True for any IP not safe to reach from a user-supplied URL. Policy: default-deny via ``ip.is_global`` (RFC 6890), plus an explicit exc

Load eval cases from a JSONL file. One JSON object per line.

Stub spend_counter_cache (and user_api_key_cache) so we can observe invalidation calls. Both caches are looked up via ``from litellm.proxy.pr

Helper to build a minimal deployment dict as returned by router.get_model_list.

Create a real httpx.HTTPStatusError for testing.

Create a mock FastAPI Request object.

Build a MagicMock prisma_client whose env-vars table returns ``row``.

Return a mock that mirrors litellm's ``AsyncHTTPHandler`` contract. Two properties of the real handler matter for these tests and were the so

Parse direct JSON response (non-streaming). Supports multiple agent response schemas: 0. {"jsonrpc": "2.0", "result": {"mess

Build a minimal ModelPreferences-like object.

Calculate response cost using result + logging object variables. used for consistent cost calculation across response headers + logg

Safely convert a 'created' field value to an integer. Some providers (like SambaNova) return the 'created' field as a float (Unix timest

Updates the config fields with the new values from the DB Args: current_config (dict): Current configuration dictionary

Write deployment health states to the router's health state cache for health-check-driven routing. No-op if the feature is disabled.

Add a success callback to litellm._async_success_callback

Classify a single user prompt. Falls back to GENERAL when no rule matches. Empty/whitespace-only also returns GENERAL.

Helper function to create a PostHog event payload for logging Args: kwargs (Dict[str, Any]): request kwargs containing s

Register a teardown action for any resource the test just created.

Check if a provider is defined via JSON

Extract original video ID without encoding.

If the request is websearch, filter out deployments that don't support web search

Snapshot, send, and drain in one consistent step. Overrides the base implementation so the same snapshot drives both the HTTP send an

Get all fallbacks for a given model from the router's fallback configuration. Args: model: The model name to get fallbacks for

Return the API base URL for the provider

Resolve xAI API keys while preserving endpoint-specific legacy order. Chat uses xai_key before XAI_API_KEY without adding a generic

Extract base URL from OpenAPI spec.

Get the cost of using built-in tools. Supported tools: - Web Search - File Search - Vector Store (Azure)

Get the guardrail translation handler for a given call type. Args: call_type: The type of call (e.g., completion, acompletion, anthr

Create a mock UserAPIKeyAuth token for testing

If model_name is provided, only return access groups for that model. Parameters: - model_name: Optional[str] - the received

Extract the original provider file/batch ID from an encoded ID. Handles IDs that start with "file-" or "batch_" prefix.

Get the set of Perplexity unified search parameters. These are the standard parameters that providers should transform from.

Extract tags from request body metadata. Args: request_body: The request body dictionary Returns: List of tag names (st

Handle callback logging failures by incrementing Prometheus metrics. Call this method in exception handlers within your callback whe

Evict from both cache layers; call after every LiteLLM_Config write.

Check if the user is allowed to call the vector store endpoint. Cover: 1. Creating a vector store index 2. Reading a vector store in