MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 12 callersMethod_scope
(self, path="/mcp", extra_headers=())
tests/unit/proxy/mcp_server/auth/test_user_api_key_auth_mcp.py:8219
↓ 12 callersFunction_scripted
(*, images: int = 1, usage: bool = True, delay: float = 0)
tests/integration/spend/test_vertex_image_regional_uplift.py:78
↓ 12 callersFunction_scripted_openai
(client_secret: Mapping[str, JsonValue])
tests/integration/providers/test_realtime_webrtc_http.py:76
↓ 12 callersFunction_search_request
Engine/app-mode search request (vertex_engine_id set).
tests/unit/llms/vertex_ai/test_vertex_ai_search_vector_store_transformation.py:151
↓ 12 callersFunction_secret_fields
(user_id: str, github_token: str)
tests/unit/llms/github_copilot/test_per_user_auth.py:629
↓ 12 callersFunction_secret_request
()
tests/unit/proxy/utils/proxy_logging/test_pre_call_hook.py:543
↓ 12 callersMethod_seed
(self, params, messages, tools=None)
tests/unit/integrations/test_anthropic_cache_control_hook.py:3189
↓ 12 callersFunction_seed_target
Raw-seed the resources target team; returns its organization_id.
tests/integration/management/test_team_update.py:64
↓ 12 callersFunction_sent_messages
(request: Request)
tests/integration/providers/test_hosted_vllm_reasoning_content_wire.py:70
↓ 12 callersFunction_serialize_streaming_chunk
(chunk: BaseModel)
litellm/proxy/proxy_server.py:10486
↓ 12 callersMethod_set_latency_metrics
( self, kwargs: dict, model: str | None, user_api_key: str | None, use
litellm/integrations/prometheus.py:2636
↓ 12 callersFunction_settled
Per trace, which of its calls reached the team's space, once every worker has sent what it decided and ``holds`` is true of every trace.
tests/integration/observability/test_otel_team_sampling_across_pods.py:295
↓ 12 callersMethod_should_raise_guardrail_blocked_exception
Only raise exception for "BLOCKED" actions, not for "ANONYMIZED" actions. If `self.mask_request_content` or `self.mask_response_cont
litellm/proxy/guardrails/guardrail_hooks/bedrock_guardrails.py:2487
↓ 12 callersFunction_single_row
(key: str, identities: tuple[str, ...])
tests/integration/spend/test_vertex_regional_uplift_spend.py:170
↓ 12 callersFunction_small_and_big
( make_core: MakeCore, small: int = 5, **arguments: object )
tests/unit/test_router/core/test_construction_precall_characterization_precall.py:81
↓ 12 callersFunction_spend_on
( limiter: PROXY_VirtualKeyModelMaxBudgetLimiter, model: str, model_max_budget: Mapping[str, object] )
tests/unit/proxy/hooks/test_model_max_budget_limiter.py:576
↓ 12 callersFunction_sse_event
(payload: dict)
tests/unit/responses/test_streaming_iterator.py:50
↓ 12 callersFunction_sse_guardrail
(**kwargs: object)
tests/unit/proxy/guardrails/guardrail_hooks/test_bedrock_guardrails.py:3144
↓ 12 callersFunction_stored_value
Pull the credential_b64 value passed to the most recent upsert call.
tests/unit/proxy/mcp_server/test_db_credentials.py:440
↓ 12 callersFunction_stream
( rig: Rig, endpoint: Endpoint, spelling: Spelling, variant: Variant, action: Action,
tests/integration/streaming/test_noma_v2_streaming_settings.py:541
↓ 12 callersFunction_streamed_reply
(text: str)
tests/integration/providers/test_ollama_prompt_tools_wire.py:95
↓ 12 callersFunction_streaming_logging_obj
()
tests/unit/proxy/test_proxy_logging_hook_detection.py:224
↓ 12 callersMethod_strip_base64_from_messages
Removes or redacts base64-encoded file data (e.g., PDFs, images, audio) from messages and responses before sending to SQS. B
litellm/integrations/custom_logger.py:996
↓ 12 callersFunction_submit
( rig: Rig, lines: Lines, key_fields: Mapping[str, JsonValue], *, finished: bool = True, token: str | None
tests/integration/observability/test_batch_line_items_cold_storage.py:432
↓ 12 callersFunction_success_kwargs
( *, model_group, deployment_model=None, response_cost=0.5, key_hash=None, key_model_m
tests/unit/proxy/hooks/test_unit_test_max_model_budget_limiter.py:606
↓ 12 callersFunction_success_line
(custom_id: str)
tests/integration/observability/test_batch_line_items_cold_storage.py:310
↓ 12 callersFunction_system_input_item
(text: str)
tests/unit/completion_extras/litellm_responses_transformation/test_completion_extras_litellm_responses_transformation_transformation.py:5838
↓ 12 callersFunction_tag
(text: str)
tests/integration/providers/test_malformed_tools_wire.py:100
↓ 12 callersFunction_tag
(label: str)
tests/integration/authorization/test_websocket_rejection_logging.py:91
↓ 12 callersFunction_team_list_caller_identity
( user_id: str | None = "caller", user_email: str | None = None, )
tests/unit/proxy/management_endpoints/test_team_endpoints.py:5244
↓ 12 callersFunction_team_list_prisma
( membership_rows: Sequence[LiteLLM_TeamMembership], budget_rows: Sequence[LiteLLM_BudgetTableFull], )
tests/unit/proxy/management_endpoints/test_team_endpoints.py:5219
↓ 12 callersFunction_text_cost
(model: str, *lines: str)
tests/unit/litellm_core_utils/test_token_counter.py:2823
↓ 12 callersFunction_tier_config
(**overrides: object)
tests/unit/router_strategy/test_complexity_router.py:14565
↓ 12 callersFunction_token
(metadata=None, team_metadata=None)
tests/unit/proxy/test_tools_allowlist_enforcement.py:21
↓ 12 callersFunction_tool_call
(index: int, arguments: str, call_id: str = None, name: str = None)
tests/unit/llms/anthropic/messages/adapters/test_streaming_iterator_tool_args.py:384
↓ 12 callersFunction_tool_msg
(content, tool_call_id="call_1")
tests/unit/litellm_core_utils/prompt_templates/test_litellm_core_utils_prompt_templates_common_utils.py:365
↓ 12 callersFunction_tool_usage_prisma
()
tests/unit/proxy/db/test_db_spend_update_writer.py:170
↓ 12 callersFunction_trace_ids
Fixed RFC 4122 shaped trace ids, like the uuid4 hex a caller mints for ``traceparent``.
tests/integration/observability/test_otel_team_sampling_across_pods.py:320
↓ 12 callersFunction_traced_app
(monkeypatch, tracer_provider)
tests/unit/integrations/otel/test_otel_v2_mount.py:313
↓ 12 callersFunction_transcript
(backend: SpeechStreamingBackend)
tests/unit/llms/vertex_ai/audio_transcription/test_vertex_ai_realtime_backend.py:119
↓ 12 callersFunction_unique_messages
()
tests/unit/caching/test_caching_handler.py:2714
↓ 12 callersFunction_update_database_with
( db_writer: DBSpendUpdateWriter, prisma: MagicMock, payload: dict, disable_spend_logs: bool =
tests/unit/proxy/db/test_db_spend_update_writer.py:4445
↓ 12 callersFunction_upload_batch_file
(headers: dict[str, str], form: dict[str, str])
tests/unit/proxy/openai_files_endpoint/test_files_endpoint.py:781
↓ 12 callersFunction_upstream
()
tests/integration/providers/test_responses_bridge_verbosity_wire.py:70
↓ 12 callersFunction_user
(marker: str, *parts: JsonValue)
tests/integration/providers/test_responses_bridge_input_audio_wire.py:91
↓ 12 callersFunction_vendor
(response_ids: Sequence[str], stream: bool)
tests/integration/providers/test_gemini_tool_result_part_reference_wire.py:165
↓ 12 callersFunction_wait_until_every_worker_serves
(gateway: Gateway, *models: str)
tests/integration/security/test_files_batches_access.py:193
↓ 12 callersFunction_wif_credential
(name: str = "federated-cred", source: CredentialSource = "db")
tests/unit/proxy/credential_endpoints/test_endpoints.py:1162
↓ 12 callersFunction_wire_batcher_for_test
Wire prisma_client.db.batch_() to return a mock batcher whose .commit() is awaitable and whose per-table .update()/.update_many() calls get c
tests/unit/proxy/common_utils/test_reset_budget_job.py:3738
↓ 12 callersFunction_with_inbound
(token: str)
tests/unit/proxy/mcp_server/outbound_credentials/test_resolver.py:107
↓ 12 callersFunctiona
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/0dvgjd47fte1d.js:1
↓ 12 callersMethodaapply
``apply``, then the model info refresh, which needs I/O.
litellm/router_rust/core.py:2105
↓ 12 callersMethodaclose
(self)
litellm/litellm_core_utils/streaming_handler.py:640
↓ 12 callersMethodacompletion
( self, model: str, messages: list, api_base: str, custom_prompt_dict:
litellm/llms/custom_llm.py:88
↓ 12 callersFunctionacompletion_with_mcp
Async completion with MCP integration. This function handles MCP tool integration following the same pattern as aresponses_api_with_mcp.
litellm/responses/mcp/chat_completions_handler.py:102
↓ 12 callersFunctionactiveTierName
(row: TierRow)
ui/litellm-dashboard/src/components/add_model/tier_rows.ts:45
↓ 12 callersMethodadd
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 12 callersMethodadd_litellm_success_callback
(self, callback: object)
tests/unit/test_router/core/scripted_host.py:207
↓ 12 callersFunctionadd_missing_team_member
- Get missing teams (diff b/w user_info.team_ids and sso_teams) - Drop the IdP groups that are not teams with one lookup, since group claims
litellm/proxy/management_endpoints/ui_sso.py:1925
↓ 12 callersFunctionadd_provider_affinity_header
( headers: Mapping[str, object], litellm_params: object | None, default_header: str | None = None )
litellm/litellm_core_utils/provider_affinity.py:141
↓ 12 callersMethodadd_update
Enqueue an update.
litellm/proxy/db/db_transaction_queue/window_spend_update_queue.py:151
↓ 12 callersMethodadd_vector_store_to_registry
Add a vector store to the registry Only add the vector store if it is not already in the registry
litellm/vector_stores/vector_store_registry.py:493
↓ 12 callersMethodafter
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 12 callersFunctionafter_text
(marker: str, hop: str, text: str, attempt: int = 0)
tests/integration/streaming/test_mid_stream_fallback_continuation.py:215
↓ 12 callersFunctionallowed_routes_check
Check if user -> not admin - allowed to access these routes
litellm/proxy/auth/auth_checks.py:1669
↓ 12 callersFunctionanthropic_deployment
(name: str, api_base: str, **fields: JsonValue)
tests/integration/providers/_cache_control_marks_support.py:449
↓ 12 callersMethodanthropic_messages_handler
LLM HTTP Handler for Anthropic Messages
litellm/llms/custom_httpx/llm_http_handler.py:2891
↓ 12 callersMethodapply_guardrail
( self, inputs: GenericGuardrailAPIInputs, request_data: dict, # mutable-ok: required
litellm/proxy/guardrails/guardrail_hooks/singulr/singulr.py:417
↓ 12 callersMethodapply_guardrail
( self, inputs: GenericGuardrailAPIInputs, request_data: dict, input_type: Lit
litellm/proxy/guardrails/guardrail_hooks/onyx/onyx.py:109
↓ 12 callersFunctionasText
(value: unknown)
ui/litellm-dashboard/src/app/(dashboard)/guardrails/_components/GuardrailFormField.tsx:52
↓ 12 callersFunctionassert_budget_refusal
(response: httpx.Response)
tests/integration/spend/_partial_stream.py:171
↓ 12 callersFunctionassert_compacted_reply
(response: httpx.Response, model: str, text: str)
tests/integration/authorization/_responses_compact.py:329
↓ 12 callersFunctionassert_payment_required_row
(call_id: str, deployment_id: str)
tests/integration/routing/test_payment_required_mapping.py:122
↓ 12 callersFunctionassert_provider_typed_rate_limit
( error: Mapping[str, JsonValue], provider_message: str = RATE_LIMIT_MESSAGE )
tests/integration/_support/responses_stream.py:208
↓ 12 callersFunctionassert_refused
( response: httpx.Response, descriptor: str, group: str, *, kind: str = "requests", limit: int = 1 )
tests/integration/routing/test_per_model_rate_limit_fallbacks.py:287
↓ 12 callersFunctionassert_stream_parity
( baseline: StreamOutcome, candidate: StreamOutcome, *, normalize: ChunkNormalizer = normalize
tests/rust-python-harness/shared/parity/stream.py:145
↓ 12 callersFunctionassertion_from_saml_login
The carrier for a SAML login: the signed ``<saml:Assertion>`` element the IdP issued, base64 encoded the way RFC 8693 carries a ``saml2`` subject
litellm/proxy/mcp_server/outbound_credentials/sso_assertion_store.py:189
↓ 12 callersMethodasync_filter_deployments
(self, *args, **kwargs)
tests/unit/test_router/test_router.py:12531
↓ 12 callersMethodasync_pre_call_deployment_hook
Pre-call hook to convert native Anthropic web_search tools to regular tools. This prevents Bedrock from trying to execute web search
litellm/integrations/websearch_interception/handler.py:446
↓ 12 callersMethodasync_pre_call_hook
Filter tools before LLM call based on user query. This hook is called before the LLM request is made. It filters the tools l
litellm/proxy/hooks/mcp_semantic_filter/hook.py:303
↓ 12 callersMethodasync_register_script
Register a Lua script with Redis asynchronously. Works with both standalone Redis and Redis Cluster. The returned callable n
litellm/caching/redis_cache.py:975
↓ 12 callersFunctionasync_safe_get
Async version of safe_get.
litellm/litellm_core_utils/url_utils.py:505
↓ 12 callersMethodasync_scan_request_attachments
(self, data: dict, call_type: CallTypesLiteral)
tests/unit/proxy/guardrails/guardrail_hooks/unified_guardrails/test_unified_guardrail.py:3264
↓ 12 callersMethodasync_set_cache
( self, key: str, value: object, litellm_parent_otel_span: Span | None,
litellm/proxy/utils.py:514
↓ 12 callersMethodattempt
(self, key: str)
litellm/proxy/common_utils/registry_read_through.py:91
↓ 12 callersFunctionbatch_cost_is_final
Whether this retrieve of the batch is the one to account its cost from. A batch still in flight has nothing to price, and a "completed" batch can
litellm/litellm_core_utils/batch_utils.py:348
↓ 12 callersMethodbatch_get_cache
( self, keys: list, parent_otel_span: Span | None = None, local_only: bool = F
litellm/caching/dual_cache.py:210
↓ 12 callersMethodbidirectional_forward
(self)
litellm/litellm_core_utils/realtime_streaming.py:2439
↓ 12 callersFunctionbind_budget_reservation_to_callbacks
Mark the request's budget reservation as owned by the success callbacks of this call. The proxy releases any reservation still unbound when the r
litellm/litellm_core_utils/core_helpers.py:627
↓ 12 callersMethodblock_on
(&self, future: impl Future<Output = T>)
litellm-rust/crates/cache-azure-blob/src/cache.rs:158
↓ 12 callersFunctionbody_of
(request: Request)
tests/integration/providers/_mantle_gpt_prompt_cache_support.py:373
↓ 12 callersFunctionbounded_db_lookup
( lookup: Awaitable[LookupT], *, name: str, deadline_seconds: float | None = None, tracker
litellm/proxy/db/db_lookup_gate.py:65
↓ 12 callersFunctionbuild_openai_client
( api_key: str | None, api_base: str | None, timeout: float | httpx.Timeout | None | NotGiven = NO
litellm/llms/openai/workload_identity.py:170
↓ 12 callersFunctionbuild_owner_filter
Return a Prisma `where` fragment that scopes a managed-resource listing to records the caller is allowed to see. - ``{}`` means no scoping (p
litellm/llms/base_llm/managed_resources/isolation.py:69
↓ 12 callersMethodbuild_request
( self, method: str, url: str, *, headers: Mapping[str, str],
tests/unit/proxy/agent_endpoints/test_kill_switch.py:35
↓ 12 callersFunctionbuild_token_endpoint_client_auth
Return the headers and body fields that authenticate the client to the token endpoint. ``client_secret_basic`` is a confidential-client method, s
litellm/proxy/mcp_server/auth/token_endpoint_auth.py:47
↓ 12 callersMethodcalculate_usage
Calculate usage for the given chunks.
litellm/litellm_core_utils/streaming_chunk_builder_utils.py:1165
↓ 12 callersFunctioncan_modify_verification_token
Check if user has permission to modify (delete/regenerate) a verification token. Rules: - Proxy admin can modify any key - Internal
litellm/proxy/management_endpoints/key_management_endpoints.py:5605
↓ 12 callersFunctioncaptureRequestBody
( page: Page, match: { method: string; urlIncludes: string }, action: () => Promise<void>, )
tests/e2e/ui/helpers/roundTrip.ts:10
↓ 12 callersFunctioncarries_token
(endpoint: Endpoint, stream: bool)
tests/integration/providers/_gemini_long_decode_support.py:530
← previousnext →5,401–5,500 of 202,479, ranked by callers