MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 6 callersFunction_resolve_provider_from_deployment
Resolve a provider hint from the deployment's litellm_params when the user-facing model name has no provider prefix. Tries the post-alia
litellm/proxy/litellm_pre_call_utils.py:3784
↓ 6 callersFunction_resolve_request_principal
Project the resolved identity into one per-request Principal, off the key object the builder already fetched, and stamp the request network contex
litellm/proxy/auth/user_api_key_auth.py:4370
↓ 6 callersFunction_resolve_team_allowed_mcp_servers
Resolve the full set of MCP server IDs a team has access to. Combines: - Direct mcp_servers list - Servers from mcp_access_groups
litellm/proxy/management_helpers/object_permission_utils.py:449
↓ 6 callersMethod_resolve_tenant
Identify the authenticated principal the routing override belongs to. API-key auth is scoped by the hashed key. JWT (and other keyle
litellm/proxy/hooks/sensitive_data_routing.py:56
↓ 6 callersMethod_resolve_timeout
Budget for one call to the compression service, unset meaning the default. Zero, negative and non-finite values are rejected instead of passe
litellm/proxy/guardrails/guardrail_hooks/headroom/headroom.py:567
↓ 6 callersFunction_responder
(backend: str)
tests/integration/providers/test_pdf_input_support_wire.py:104
↓ 6 callersFunction_response
(*, content=None, tool_calls=None, finish_reason="stop", model="gpt-4o")
tests/unit/proxy/mcp_server/test_mcp_sampling_response_conversion.py:36
↓ 6 callersFunction_response
(payload)
tests/unit/llms/mistral/files/test_mistral_files_transformation.py:39
↓ 6 callersFunction_response
(payload: dict, status_code: int = 200)
tests/unit/llms/mistral/batches/test_mistral_batches_transformation.py:56
↓ 6 callersFunction_response_asking_for_searches
(*queries: str)
tests/unit/integrations/websearch_interception/test_websearch_agentic_loop_cap.py:626
↓ 6 callersFunction_response_data
(raw_response: httpx.Response)
litellm/llms/fal_ai/videos/transformation.py:188
↓ 6 callersFunction_responses_api_response_without_usage
()
tests/unit/responses/test_streaming_iterator.py:787
↓ 6 callersFunction_responses_entry
( body: Mapping[str, object] | None = None, custom_id: str = "resp-1", url: str = "/v1/responses",
tests/unit/llms/vertex_ai/files/test_vertex_ai_files_transformation.py:1481
↓ 6 callersFunction_restricted_key
(gateway: Gateway, scenario: Scenario)
tests/integration/providers/test_openai_error_handling.py:8
↓ 6 callersFunction_retrieve_batch
(gateway: Gateway, batch: _OpenAIBatch)
tests/integration/management/test_batch_output_file_listing.py:234
↓ 6 callersFunction_retry_count_texts
(lines: tuple[str, ...])
tests/integration/database/test_v1_migration_error_logging.py:201
↓ 6 callersFunction_reuse_persisted_dcr_client_if_available
( mcp_server: MCPServer, current_redirect_uri: str | None = None )
litellm/proxy/mcp_server/discoverable_endpoints.py:1772
↓ 6 callersMethod_rewrite_defs_refs
Recursively rewrite $ref values from #/$defs/... to #/components/schemas/... This converts Pydantic v2 references to OpenAPI-compatib
litellm/proxy/common_utils/custom_openapi_spec.py:228
↓ 6 callersFunction_root_trace_name_after
( logger: OpenTelemetryV2, calls: Sequence[tuple[str, Mapping[str, object]]], root_logger: OpenTel
tests/unit/integrations/otel/test_langfuse_logger.py:451
↓ 6 callersFunction_roster_user_roles
(members_json: str | None)
tests/unit/proxy/management_endpoints/test_team_endpoints.py:15110
↓ 6 callersMethod_route
( self, router: Router, request: dict[str, object] | None = None, model: str =
tests/unit/test_router/test_router.py:13198
↓ 6 callersMethod_route_matches_pattern
Check if route matches the pattern placed in proxy/_types.py Example: - pattern: "/threads/{thread_id}" - route: "/t
litellm/proxy/auth/route_checks.py:601
↓ 6 callersFunction_routed_by
Metadata as a pre-routing strategy leaves it on the request it served.
tests/unit/integrations/test_shadow_eval_logger.py:260
↓ 6 callersMethod_routed_call_kwargs
(router: Router, prompt: str = "hi", **request_params)
tests/unit/router_strategy/test_complexity_router.py:5370
↓ 6 callersFunction_routed_entry
(name: str, api_base: str, litellm_model: str, info: Mapping[str, JsonValue])
tests/integration/routing/test_non_chat_model_endpoint_gate.py:843
↓ 6 callersFunction_router
(content_policy_fallbacks: list | None)
tests/unit/router_utils/test_fallback_event_handlers.py:2024
↓ 6 callersMethod_router
( model_id: str, litellm_params: dict[str, object], allowed_fails_policy: dict[str, in
tests/unit/router_utils/test_cooldown_handlers.py:1132
↓ 6 callersMethod_router
( model_id: str, litellm_params: Mapping[str, object], allowed_fails_policy: Mapping[s
tests/unit/router_utils/test_cooldown_handlers.py:1305
↓ 6 callersMethod_router
(self, marker_litellm_params: dict)
tests/unit/test_router/test_router.py:10530
↓ 6 callersMethod_router
(cls, *markers: dict[str, object])
tests/unit/test_router/test_router.py:13160
↓ 6 callersMethod_router
(mock_router_instance, **overrides)
tests/unit/router_strategy/test_complexity_router.py:17540
↓ 6 callersFunction_router_rejected_failure_payload
(model_group: str, llm_router: litellm.Router | None)
tests/unit/proxy/spend_tracking/test_spend_tracking_utils.py:5805
↓ 6 callersFunction_router_serving
(*model_ids: str)
tests/unit/proxy/common_utils/test_registry_read_through.py:515
↓ 6 callersFunction_router_with_io_limits
()
tests/unit/test_router/test_io_token_rate_limits.py:1303
↓ 6 callersMethod_router_with_tiers
(tiers: dict, **kwargs)
tests/unit/router_strategy/test_complexity_router.py:12491
↓ 6 callersFunction_routes
(submission_status: str, retrieve: JsonResponse)
tests/integration/spend/test_background_response_poll_retirement.py:87
↓ 6 callersFunction_routing_router
( strategy: RoutingStrategy | str, a_params: dict[str, float] | None = None, b_params: dict[str, f
tests/unit/test_router/test_router_provider_endpoints.py:413
↓ 6 callersFunction_row
(proxy: Gateway, where: str, value: str)
tests/integration/spend/test_spend_log_latest_turn.py:105
↓ 6 callersFunction_row_values
(row: Mapping[str, JsonValue])
tests/integration/spend/test_passthrough_upstream_model.py:68
↓ 6 callersFunction_rows
(model: str, call_ids: frozenset[str])
tests/integration/messages_endpoint/chat_bridge/test_chat_bridge_mid_stream_failure_bookkeeping_wire.py:164
↓ 6 callersFunction_rows
(deployments: object)
tests/unit/test_router/core/test_gaps_l14000.py:146
↓ 6 callersFunction_run
()
tests/unit/llms/anthropic/messages/responses_adapters/test_responses_adapters_streaming_iterator.py:45
↓ 6 callersMethod_run
(self, processing_obj, monkeypatch, chunks, stream=None, proxy_logging_obj=None, request=None)
tests/unit/proxy/test_common_request_processing.py:6586
↓ 6 callersFunction_run_chat_completions_pipeline
Anthropic /v1/messages input -> chat adapter -> the OpenAI-compatible request transformation every OpenAIGPTConfig-based provider runs.
tests/unit/llms/anthropic/messages/adapters/test_anthropic_messages_adapters_transformation.py:4663
↓ 6 callersMethod_run_guardrail_with_metrics
Await `coro`, recording its latency and status to the `litellm_guardrail_latency_seconds` metric under `hook_type`, and enric
litellm/proxy/utils.py:3581
↓ 6 callersFunction_run_hooks
(handler: BaseLLMHTTPHandler, logging_obj: MagicMock)
tests/unit/integrations/websearch_interception/test_websearch_streaming_wrap.py:40
↓ 6 callersFunction_run_init_coordination_redis
Run ProxyConfig._init_coordination_redis against a stubbed module state, returning (redis_usage_cache, spend_counter redis, config-cache redis).
tests/unit/proxy/test_proxy_server_endpoints_and_startup.py:13320
↓ 6 callersMethod_run_lasso_guardrail
Run the Lasso guardrail with the specified message type. This is the core method that handles both classification and masking workfl
litellm/proxy/guardrails/guardrail_hooks/lasso/lasso.py:378
↓ 6 callersFunction_run_new_project
(data: NewProjectRequest)
tests/unit/enterprise/proxy/management_endpoints/test_project_endpoints_prisma.py:1224
↓ 6 callersFunction_run_refresh
(monkeypatch, server, response_body=None)
tests/unit/proxy/mcp_server/test_db_credentials.py:1303
↓ 6 callersMethod_run_scheduled_export
Export today's window, catching up any dates Mavvrik has not yet received. On each run: 1. Register with Mavvrik → get metricsMarker
litellm/integrations/mavvrik_focus/mavvrik_focus_logger.py:162
↓ 6 callersFunction_run_setup_prisma_client
(mock_client)
tests/unit/proxy/test_proxy_server_endpoints_and_startup.py:14347
↓ 6 callersFunction_running_scheduler
A started scheduler with every job in flight, stopped on the way out whatever the test did
tests/unit/proxy/shutdown/test_scheduled_jobs.py:44
↓ 6 callersFunction_runtime_models
(wire: Wire)
tests/integration/providers/test_count_tokens_model_wire.py:196
↓ 6 callersFunction_s256
(verifier: str)
tests/unit/proxy/mcp_server/test_discoverable_endpoints.py:11985
↓ 6 callersFunction_sagemaker_router
( model: str, deployment_params: Mapping[str, object], mode: str, router_defaults: Mapping[str
tests/unit/test_router/test_router.py:20118
↓ 6 callersMethod_same_provider
Return True if model uses the same LLM provider as the connection model.
litellm/responses/streaming_iterator.py:3469
↓ 6 callersFunction_sanitize_empty_content
Remove or filter content so empty text blocks are not sent. Databricks Model Serving uses Anthropic Messages API spec and rejects empty text
litellm/llms/databricks/chat/transformation.py:223
↓ 6 callersFunction_sanitize_id
Reject path traversal characters and URL-encode the identifier.
litellm/integrations/arize/arize_phoenix_client.py:19
↓ 6 callersFunction_save_email_settings
Helper function to save email settings to general_settings in db
enterprise/litellm_enterprise/enterprise_callbacks/send_emails/endpoints.py:64
↓ 6 callersMethod_scan_text
Scan one text: find blocks, apply block/mask/allow by confidence. When detect_execution_intent is True and input_type is "request", o
litellm/proxy/guardrails/guardrail_hooks/block_code_execution/block_code_execution.py:405
↓ 6 callersFunction_scanned_prompts
(scans: tuple[Request, ...])
tests/integration/observability/test_guardrail_effects.py:526
↓ 6 callersFunction_scanned_while_upstream_is_held
( rig: _DisconnectRig, disconnect: Callable[[_DisconnectRig, int], str], index: int = 0 )
tests/integration/observability/test_guardrail_effects.py:2582
↓ 6 callersFunction_schema
(response: httpx.Response)
tests/integration/configuration/test_openapi_docs_route.py:37
↓ 6 callersMethod_scope_tag
(cls, key: str)
litellm/caching/valkey_semantic_cache.py:119
↓ 6 callersFunction_scoped
(scope, scope_id, value, source="key", setting="max_batch_file_uploads_per_day")
tests/unit/proxy/openai_files_endpoint/test_file_usage_caps.py:29
↓ 6 callersFunction_scoping_db_tools
()
tests/unit/proxy/management_endpoints/search_endpoints/test_search_tool_management.py:611
↓ 6 callersFunction_scrape
(client: httpx.AsyncClient, headers: Mapping[str, str] | None = None)
tests/unit/integrations/test_prometheus_metrics_endpoint.py:56
↓ 6 callersFunction_scripted
(reply: Reply)
tests/integration/sdk/test_provider_error_and_extra_body_families_wire.py:123
↓ 6 callersFunction_scripted
(scenario: Scenario, prefix: str, response: JsonResponse | BinaryResponse)
tests/integration/compatibility/test_status_code_followups.py:104
↓ 6 callersFunction_scripted_deployment
(scenario: Scenario, marker: str, routes: RoutedResponse)
tests/integration/spend/test_background_response_poll_retirement.py:102
↓ 6 callersFunction_sdk_base
(gateway: Gateway)
tests/integration/routing/test_complexity_router_databricks_classifier.py:446
↓ 6 callersFunction_seal
Serialized ``exclude_none`` for the same reason session JWTs are minted that way: an optional claim that is unset never reaches the wire, so durin
litellm/proxy/mcp_server/gateway_dcr_flow.py:484
↓ 6 callersFunction_sealed_refresh
(keys: EnvelopeKeys, now: datetime = _NOW, identity: EnvelopeIdentity = _IDENTITY)
tests/unit/proxy/mcp_server/outbound_credentials/test_bridge_credentials.py:62
↓ 6 callersFunction_search_response
(payload: Mapping[str, object])
tests/unit/llms/vertex_ai/test_vertex_ai_search_vector_store_transformation.py:314
↓ 6 callersFunction_search_response
()
tests/unit/integrations/websearch_interception/test_websearch_rich_query_shape.py:34
↓ 6 callersFunction_seed
(db, assignments)
tests/integration/management/test_access_group_team_sync.py:56
↓ 6 callersFunction_seed_daily_tag_spend
(tag: str, days: tuple[str, ...])
tests/integration/spend/test_daily_tag_spend_retention.py:29
↓ 6 callersFunction_seed_max_parallel_requests_slots
( handler: _PROXY_MaxParallelRequestsHandler, gauge: RateLimitGauge, slot_ids: list[str] )
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:4981
↓ 6 callersFunction_seed_settings_store
Point proxy_config.settings at a store holding the same row the mocked table returns, the way a booted proxy does, so the read routes resolve agai
tests/unit/proxy/proxy_server/test_routes_config.py:30
↓ 6 callersFunction_seed_unhealthy_states
(router, unhealthy_ids, timestamp=None)
tests/unit/test_router/test_router.py:6365
↓ 6 callersFunction_send
(rig: _Rig, path: str, body: Mapping[str, JsonValue])
tests/integration/providers/test_anthropic_empty_user_turn_wire.py:111
↓ 6 callersFunction_send
(gateway: Gateway, call: _Call, key: str, **extra: JsonValue)
tests/integration/observability/test_prometheus_shared_custom_labels.py:187
↓ 6 callersFunction_send
(hook: str, kwargs: Mapping[str, object])
tests/integration/_support/hook_count_probe.py:20
↓ 6 callersFunction_send_message
(proxy: Gateway, provider: SharedProvider, key: str)
tests/integration/authorization/test_cli_sso_login_ui_disabled.py:265
↓ 6 callersFunction_send_unmarked_system_prompt
(router: Router, host: str, entrypoint: str)
tests/unit/integrations/test_anthropic_cache_control_hook.py:5270
↓ 6 callersFunction_sent_count_body
( monkeypatch: pytest.MonkeyPatch, messages: list[dict[str, object]], system: object, messages
tests/unit/llms/openai/responses/count_tokens/test_token_counter.py:176
↓ 6 callersFunction_serve_speak
(websocket: _FakeWebSocket, user_api_key_dict: UserAPIKeyAuth | None = None)
tests/unit/proxy/pass_through_endpoints/test_deepgram_ws_passthrough_routes.py:501
↓ 6 callersFunction_served
( upstream: AsyncIterator[ModelResponseStream], request_data: dict[str, object] | None = None )
tests/unit/proxy/proxy_server/test_streaming_helpers.py:1549
↓ 6 callersMethod_server
(self, extra_headers)
tests/unit/proxy/mcp_server/test_mcp_hook_extra_headers.py:1144
↓ 6 callersMethod_server
(alias, server_name, server_id)
tests/unit/proxy/mcp_server/test_mcp_toolset_scope.py:545
↓ 6 callersMethod_server_fulfilled_tools_in_request
The request's tools that a registered callback fulfills server-side (e.g. ``headroom_retrieve``).
litellm/llms/custom_httpx/llm_http_handler.py:6018
↓ 6 callersMethod_server_grants_with_group_scope
Keep a configured empty group scope distinct from an unrestricted level. Combine same-level grants first; the existing zero-server marker pre
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:3556
↓ 6 callersFunction_server_with_auth
(auth_type)
tests/unit/proxy/mcp_server/outbound_credentials/test_sso_assertion_store.py:103
↓ 6 callersMethod_servers
(self)
tests/unit/proxy/management_endpoints/test_mcp_management_endpoints.py:8012
↓ 6 callersFunction_session
(db: Prisma, record: BaselineAccountingRecord)
tests/integration/spend/test_baseline_accounting.py:123
↓ 6 callersFunction_session_cookie_user_id
The signed-in litellm user for a browser request, or ``None``. Thin wrapper so the aggregate DCR flow's verbs receive the identity as a plain valu
litellm/proxy/mcp_server/discoverable_endpoints.py:476
↓ 6 callersFunction_session_user_row
(*, teams=(SESSION_TEAM_ID,), role=LitellmUserRoles.INTERNAL_USER, models=())
tests/unit/proxy/auth/test_user_api_key_auth_request_flow.py:9095
↓ 6 callersMethod_set_customer_budget_metrics_after_api_request
( self, end_user_id: str | None, response_cost: float, )
litellm/integrations/prometheus.py:4625
↓ 6 callersMethod_set_org_budget_metrics
Set org budget metrics for a single org - Remaining Budget - Max Budget - Budget Reset At
litellm/integrations/prometheus.py:4575
← previousnext →12,301–12,400 of 202,479, ranked by callers