MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 15 callersFunctione
()
litellm/proxy/_experimental/out/_next/static/chunks/0dvgjd47fte1d.js:1
↓ 15 callersFunctione
(t)
litellm/proxy/_experimental/out/_next/static/chunks/0cz1d0mv5g_q7.js:1
↓ 15 callersFunctioneffective_skip_system_message_for_guardrail
(guardrail_to_apply: object)
litellm/llms/base_llm/guardrail_translation/utils.py:230
↓ 15 callersMethodelapsed
(self)
litellm/llms/custom_httpx/aiohttp_pool_diagnostics.py:54
↓ 15 callersFunctionensure_request_log_indexes
Build every listed index that is missing or invalid. Each build step runs under the migration coordinator lock, held per statement so a resolver b
litellm-proxy-extras/litellm_proxy_extras/request_log_indexes.py:155
↓ 15 callersFunctionenter_native
Claims process-bound native state before runtime startup or tokenizer execution.
litellm-rust/crates/host-python/src/runtime.rs:45
↓ 15 callersFunctionerror_body
(response: httpx.Response)
tests/integration/_support/responses_stream.py:173
↓ 15 callersFunctioneu
(e)
litellm/proxy/_experimental/out/_next/static/chunks/3lb5gjzjxjxem.js:1
↓ 15 callersMethodexecute
(self, operation: DiscoverRequest, context: OperationContext)
litellm/proxy/mcp_server/operations.py:3785
↓ 15 callersFunctionextract_compression_saved_tokens
Return the total prompt tokens saved by compression for one request. Sums two disjoint sources: - the native ``compression_savings`` ke
litellm/proxy/spend_tracking/compression_savings.py:42
↓ 15 callersFunctionextract_user_id_from_request
Resolve the caller for identity binding without granting credential-write permission.
litellm/proxy/mcp_server/bridge_token_flow.py:437
↓ 15 callersMethodfetch
(self, user_id: str)
tests/unit/proxy/mcp_server/outbound_credentials/test_resolver.py:698
↓ 15 callersMethodfind_team_with_model_access
Find first team with access to the requested model
litellm/proxy/auth/handle_jwt.py:1755
↓ 15 callersMethodflush
Flush the logging queue. Waits until every enqueued task has completed. ``queue.join()`` blocks on the queue's unfinished-task counte
litellm/litellm_core_utils/logging_worker.py:531
↓ 15 callersMethodflush
({getSearchParamsSnapshot:e=c,rateLimitFactor:r=1,...s},n)
litellm/proxy/_experimental/out/_next/static/chunks/0tx19jeq-wskb.js:2
↓ 15 callersMethodfrom_tiktoken
(encoding: &str)
litellm-rust/crates/token-counter/src/tiktoken.rs:6
↓ 15 callersFunctiongenerate_key_helper_fn
( request_type: Literal["user", "key"], # identifies if this request is from /user/new or /key/generate
litellm/proxy/management_endpoints/key_management_endpoints.py:5248
↓ 15 callersFunctiongeneric_response_convertor
( response, jwt_handler: JWTHandler, sso_jwt_handler: JWTHandler | None = None, role_mappings:
litellm/proxy/management_endpoints/ui_sso.py:1263
↓ 15 callersFunctionget_all_fallbacks
Get all fallbacks for a given model from the router's fallback configuration. Args: model: The model name to get fallbacks for
litellm/proxy/auth/model_checks.py:867
↓ 15 callersMethodget_available_deployment_for_pass_through
Deprecated: ``select(model, kind="pass_through")``.
litellm/router_rust/core.py:2538
↓ 15 callersMethodget_cache
(self, key: str)
litellm/proxy/auth/login_throttle.py:72
↓ 15 callersMethodget_cache
(self, key: str)
tests/unit/router_strategy/test_least_busy.py:588
↓ 15 callersMethodget_config_state
Returns a deep copy of the config, Do this, to avoid mutating the config state outside of allowed methods
litellm/proxy/proxy_server.py:6640
↓ 15 callersFunctionget_cost_per_unit
(model_info: ModelInfo, cost_key: str, default_value: float | None = 0.0)
litellm/litellm_core_utils/llm_cost_calc/utils.py:923
↓ 15 callersMethodget_data
(self, *, user_id: str)
litellm/proxy/management_endpoints/internal_user_endpoints.py:836
↓ 15 callersFunctionget_deployment_failures_for_current_minute
Returns the number of fails for the current minute for a deployment_id Returns 0 if no value found
litellm/router_utils/router_callbacks/track_deployment_metrics.py:98
↓ 15 callersFunctionget_deployments_for_tag
Returns a list of deployments that match the requested model and tags in the request. Executes tag based filtering based on the tags in requ
litellm/router_strategy/tag_based_routing.py:560
↓ 15 callersMethodget_error_class
Return CometAPI-specific error class
litellm/llms/cometapi/chat/transformation.py:125
↓ 15 callersMethodget_levo_config
Retrieves the Levo configuration based on environment variables. Returns: LevoConfig: Configuration object containing Le
litellm/integrations/levo/levo.py:37
↓ 15 callersMethodget_litellm_managed_vector_store_from_registry
Returns the vector store from the registry
litellm/vector_stores/vector_store_registry.py:250
↓ 15 callersFunctionget_litellm_user_role
Convert a string (or list of strings) to a LitellmUserRoles enum if valid (case-insensitive). Handles list inputs since some SSO providers (
litellm/proxy/management_endpoints/types.py:71
↓ 15 callersFunctionget_non_default_completion_params
(kwargs: Mapping[str, object])
litellm/utils.py:11896
↓ 15 callersMethodget_openai_compatible_provider_info
( self, api_base: str | None, api_key: str | None, litellm_params: GenericLite
litellm/llms/bedrock_mantle/chat/transformation.py:139
↓ 15 callersMethodget_prompt
Fetch a prompt definition from the MCP server.
litellm/mcp_client/client.py:1369
↓ 15 callersFunctionget_protected_indices
Return indices of messages that must never be compressed: - All system messages - The last user message - The last assistant message
litellm/compression/compress.py:225
↓ 15 callersMethodget_provider_video_config
( model: str | None, provider: LlmProviders | str, )
litellm/utils.py:11265
↓ 15 callersMethodget_queue
Return a queue for that specific model group. Redis hands the queue back as JSON lists, so every entry is validated into the
litellm/scheduler.py:172
↓ 15 callersFunctionget_spend_logs_id
( call_type: str, response_obj: Mapping[str, object], kwargs: Mapping[str, object], metadata:
litellm/proxy/spend_tracking/spend_tracking_utils.py:373
↓ 15 callersFunctiongranted_toolset_ids
Toolset ids the principal holds, resolved per grant source with the key/team rule the aggregate /mcp listing applies: a source that declares any M
litellm/proxy/mcp_server/ui_session_utils.py:259
↓ 15 callersFunctionh
(t,s)
litellm/proxy/_experimental/out/_next/static/chunks/3yrgf44a9g52e.js:7
↓ 15 callersMethodhandle_count_tokens_request
( self, request_data: dict[str, object], litellm_params: dict[str, object], re
tests/unit/proxy/test_proxy_token_counter.py:1007
↓ 15 callersFunctionhashed
(key: str)
tests/integration/management/_batch_file_caps.py:316
↓ 15 callersFunctioninference_request_view
Apply the history rewrites and context edits /v1/messages runs before it converts a body for another provider.
litellm/llms/anthropic/messages/context_management/inference_view.py:125
↓ 15 callersMethodinfo
(model, **kwargs)
tests/unit/proxy/mcp_server/test_mcp_sampling_priority_selection.py:180
↓ 15 callersFunctioninput
(pieces: Vec<Bytes>)
litellm-rust/crates/framer/tests/support/mod.rs:45
↓ 15 callersMethodint
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2j7hjtph8n74b.js:1
↓ 15 callersMethodinto_json
(self)
litellm-rust/crates/llms-types/src/formats/ocr/response.rs:90
↓ 15 callersFunctioninvalidate_config_param
Evict from both cache layers; call after every LiteLLM_Config write.
litellm/proxy/utils.py:5541
↓ 15 callersFunctionisIterable
(s)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 15 callersMethodis_empty
(&self)
litellm-rust/crates/cache-memory/src/cache.rs:168
↓ 15 callersFunctionis_otel_v2_enabled
()
litellm/integrations/otel/model/config.py:56
↓ 15 callersMethodjsonify_object
(self, data: Mapping[str, object])
litellm/proxy/utils.py:5906
↓ 15 callersFunctionlisting
( fetch, cursor: str | None = None, *, caller_scope: str = "caller-and-scope", snapshot: s
tests/unit/proxy/mcp_server/test_catalog.py:215
↓ 15 callersMethodlocal
(self, entries: Sequence[RateLimitEntry])
litellm/redis/stores/rate_limits/__init__.py:87
↓ 15 callersFunctionlock
(mutex: &Mutex<T>)
litellm-rust/crates/redis-store/src/memory.rs:120
↓ 15 callersFunctionlogging_safe_mcp_headers
The MCP request's client headers, sanitized the way the chat completions path sanitizes them before they reach a logging callback or a guardrail:
litellm/proxy/mcp_server/utils.py:1099
↓ 15 callersFunctionmake_batch
( *, id: str = "batch-provider-id", output_file_id: Optional[str] = None, error_file_id: Optio
tests/unit/proxy/batches_endpoints/test_endpoints.py:94
↓ 15 callersFunctionmake_context
( tmp_path: Path, *, model: str | None = "gpt-4o-mini", gateway: GatewayTarget | None = None,
tests/unit/harness/handlers/test_tool_loop_handler.py:71
↓ 15 callersFunctionmake_ctx
(sandbox: FakeSandbox, **overrides: Any)
tests/unit/llms/claude_code/harness/test_transformation.py:185
↓ 15 callersFunctionmake_handler
(completion: ScriptedCompletion)
tests/unit/harness/handlers/test_tool_loop_handler.py:102
↓ 15 callersFunctionmake_router
( window: int | None = 512, settings: Mapping[str, object] | None = None, *, compactor_window: int = 32000
tests/unit/router_strategy/complexity_router/test_context_compaction.py:41
↓ 15 callersFunctionmake_tool_call_dict
Create a tool call dict matching the ChatCompletionMessageToolCall schema.
tests/unit/integrations/rubrik_test_helpers.py:8
↓ 15 callersFunctionmanaged_rows
(response_id: str)
tests/integration/spend/_background_responses.py:289
↓ 15 callersMethodmethod
()
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 15 callersMethodmodel_dump
(self)
litellm/proxy/container_endpoints/ownership.py:116
↓ 15 callersMethodmodel_dump
(self, *, exclude_none: bool)
litellm/integrations/mlflow.py:22
↓ 15 callersFunctionnew_key
(rig: Rig)
tests/integration/streaming/test_mid_stream_fallback_continuation.py:381
↓ 15 callersFunctionnormalize_server_name
Normalize server name by replacing spaces with underscores
litellm/proxy/mcp_server/utils.py:165
↓ 15 callersMethodobserve_client_event
(self, event: Mapping[str, object])
litellm/llms/xai/realtime/usage_meter.py:75
↓ 15 callersMethodopen_persistent_session
( self, admission: Callable[[], contextlib.AbstractAsyncContextManager[None]] | None = None,
tests/unit/proxy/mcp_server/test_mcp_server_manager.py:19282
↓ 15 callersMethodopenai_passthrough_handler
Handle OpenAI passthrough logging with cost tracking for chat completions, embeddings, image generation, image editing, and responses
litellm/proxy/pass_through_endpoints/llm_provider_handlers/openai_passthrough_logging_handler.py:476
↓ 15 callersFunctionoperation
(session: ClientSession)
tests/unit/mcp_client/test_mcp_client.py:2266
↓ 15 callersMethodpayloads
(self)
tests/integration/observability/_s3_v2_support.py:80
↓ 15 callersMethodpeek
(&self)
litellm-rust/crates/python-compat/src/literal.rs:78
↓ 15 callersMethodping
(&self)
litellm-rust/crates/redis-client/src/client.rs:79
↓ 15 callersFunctionplan_for
(endpoint: str)
litellm/router_rust/invoke/endpoints.py:693
↓ 15 callersMethodposition
`list[value]` as Python reads it: an int (or bool) in range, negative from the end.
litellm-rust/crates/router-core/src/lookup.rs:118
↓ 15 callersMethodpre_call_hook
( self, *, user_api_key_dict: UserAPIKeyAuth, data: dict[str, object], # muta
litellm/proxy/response_api_endpoints/endpoints.py:2181
↓ 15 callersMethodprocess_output_response
Scan the text content of an MCP tool result and write masked text back. The content list is rewritten in place (only the entries the guardrai
litellm/proxy/mcp_server/guardrail_translation/handler.py:228
↓ 15 callersMethodprocess_rotations
Main entry point - find and rotate keys that are due for rotation. Uses PodLockManager to ensure only one pod runs rotation in multi-
litellm/proxy/common_utils/key_rotation_manager.py:49
↓ 15 callersMethodquery
(self, content: str, *, tx_id: str | None)
tests/unit/proxy/db/test_prisma_client.py:517
↓ 15 callersFunctionreceived_voice
(request: Request)
tests/integration/providers/_gemini_tts_voice_support.py:103
↓ 15 callersMethodreplayed
(self)
tests/_ws_vcr.py:436
↓ 15 callersFunctionrequiredRule
(message: string)
ui/litellm-dashboard/src/components/common_components/formRules.ts:23
↓ 15 callersFunctionresolve
(s,o,i)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 15 callersFunctionresolve_litellm_call_id
(client_call_id: str | None)
litellm/proxy/common_request_processing.py:379
↓ 15 callersFunctionresponse
(model: str = "claude-sonnet-4-5")
tests/test_litellm_rust/messages/test_dispatch.py:43
↓ 15 callersMethodresume
Hand back the response to the pending host call (`None` on the first call) and run the call to its next step or its end.
litellm-rust/crates/python-bridge/src/router/host.rs:1095
↓ 15 callersFunctionroute_a2a_agent_request
Route A2A agent requests directly to litellm with injected API base. Returns None if not an A2A request (allows normal routing to continue).
litellm/proxy/agent_endpoints/a2a_routing.py:123
↓ 15 callersMethodrouter_model_info
Model info used for routing decisions (was ``get_router_model_info``).
litellm/router_api.py:400
↓ 15 callersMethodrun
(self)
tests/integration/_support/bedrock_realtime_peer.py:417
↓ 15 callersFunctionrun_async_value
(py: Python<'_>, future: F)
litellm-rust/crates/python-bridge/src/execution.rs:40
↓ 15 callersFunctionrun_off_loop_if_aws
( provider_config: object, call: Callable[_SignParams, _SignedRequest], /, *args: _SignParams.
litellm/llms/bedrock/base_aws_llm.py:1750
↓ 15 callersMethodsend
( self, key: str, endpoint: Endpoint = "chat", stream: bool = False, *
tests/integration/observability/test_otel_team_capture_message_content.py:230
↓ 15 callersMethodsetSelection
(e,t,n,s)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 15 callersMethodset_attribute
(self, key, value)
tests/unit/integrations/otel/test_otel_v2_mount.py:107
↓ 15 callersMethodset_cache
(self, key, value, **kwargs)
tests/unit/router_utils/test_health_state_cache.py:158
↓ 15 callersMethodset_cache
(self, key: str | None, value: object, local_only: bool = False, **kwargs: object)
tests/unit/proxy/proxy_server/test_spend_counters.py:1651
↓ 15 callersFunctionsettle_chat
( rig: AliasRig, model: str, key: str, status: int, *, seconds: float = 60, extra:
tests/integration/authorization/_hidden_alias_budget.py:186
↓ 15 callersMethodsetup_database
Set up the database using either prisma migrate or prisma db push Uses migrations from litellm-proxy-extras package The requ
litellm-proxy-extras/litellm_proxy_extras/utils.py:1400
← previousnext →4,101–4,200 of 202,479, ranked by callers