MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 6 callersFunction_prompted
(wire: Wire)
tests/integration/sdk/test_provider_error_and_extra_body_families_wire.py:127
↓ 6 callersMethod_propagate_usage_cost_to_hidden_params
The builder's usage only keeps declared fields, so provider blocks like usage.opencode are read off the chunks.
litellm/litellm_core_utils/streaming_handler.py:2263
↓ 6 callersFunction_provider_calls
(wire: Wire)
tests/integration/providers/test_hosted_vllm_reasoning_content_wire.py:89
↓ 6 callersFunction_provider_file_route
(provider: str)
tests/unit/enterprise/proxy/hooks/test_managed_files.py:2464
↓ 6 callersFunction_provider_turns
(rig: _Rig, marker: str)
tests/integration/providers/test_anthropic_empty_user_turn_wire.py:118
↓ 6 callersFunction_proxy_over
(prisma_client: MagicMock, proxy_logging: MagicMock, router: MagicMock)
tests/unit/proxy/response_api_endpoints/test_endpoints.py:1415
↓ 6 callersFunction_proxy_root
()
tests/integration/_support/process.py:706
↓ 6 callersFunction_proxy_router_with_two_model_groups
()
tests/unit/proxy/test_route_llm_request.py:1095
↓ 6 callersFunction_public_settings
(settings: ROISettings)
litellm/proxy/management_endpoints/roi_calculator_endpoints.py:204
↓ 6 callersFunction_pull
( number: int = 42, emails: tuple[str, ...] | None = None, estimate_status: Literal["estimated", "
tests/unit/proxy/roi_calculator/test_analytics.py:16
↓ 6 callersFunction_pull
(login: str, number: int = 1, repo: str = "org/service", **fields: object)
tests/unit/proxy/roi_calculator/test_observed_analytics.py:18
↓ 6 callersFunction_put
(store: _ReservationStore, key: str, values: tuple[object, ...], ttl: float)
litellm/router_strategy/least_busy.py:235
↓ 6 callersMethod_query
(url: str)
tests/unit/litellm_proxy_extras/test_litellm_proxy_extras_utils.py:1580
↓ 6 callersFunction_query_raw
(prisma_client: "PrismaClient", query: str, *args: object)
litellm/proxy/management_endpoints/auto_router_endpoints.py:238
↓ 6 callersFunction_question
(marker: str)
tests/integration/messages_endpoint/providers/bedrock/test_bedrock_messages_gpt_chat_completions_wire.py:19
↓ 6 callersFunction_question
(marker: str)
tests/integration/routing/test_deployment_mode_routing_wire.py:82
↓ 6 callersMethod_queue
(self, logger, extra)
tests/unit/integrations/test_langsmith_init.py:252
↓ 6 callersMethod_queue_final_tool_call_done_events
Ensure tool calls that were not streamed as deltas still get emitted before response.completed.
litellm/responses/litellm_completion_transformation/streaming_iterator.py:527
↓ 6 callersFunction_quota_router
(public_name: str, proxy_params: Mapping[str, object], *team_params: Mapping[str, object])
tests/unit/proxy/management_endpoints/test_model_management_endpoints.py:8277
↓ 6 callersFunction_raise_and_catch
(exc)
tests/unit/litellm_core_utils/test_litellm_logging.py:9781
↓ 6 callersMethod_raise_for_response
( response: httpx.Response, provider_config: SonioxAudioTranscriptionConfig, action: s
litellm/llms/soniox/audio_transcription/handler.py:369
↓ 6 callersFunction_raise_if_ptu_share_teams_missing
Hold every team named in ``ptu_shares`` to the existence check ``team_id`` already gets.
litellm/proxy/management_endpoints/model_management_endpoints.py:1122
↓ 6 callersFunction_rate_limited_error
(retry_after: str | None = None)
tests/unit/proxy/mcp_server/faults/test_list_outcomes.py:394
↓ 6 callersFunction_raw
(gateway: Gateway, path: str, body: Mapping[str, JsonValue])
tests/integration/providers/test_responses_bridge_tool_search_replay_wire.py:209
↓ 6 callersFunction_raw_call
( client: httpx.AsyncClient, path: str, body: Mapping[str, JsonValue], call: _Call, )
tests/integration/providers/_responses_bridge_prompt_cache_breakpoint.py:402
↓ 6 callersFunction_raw_image_response
(mock_image_data)
tests/unit/llms/openai/image_generation/test_openai_image_generation_extra_headers.py:18
↓ 6 callersFunction_read
(db)
tests/integration/management/test_access_group_team_sync.py:67
↓ 6 callersFunction_read
(gateway: Gateway, path: str)
tests/integration/routing/test_prompt_cache_pin_memory_routes.py:81
↓ 6 callersFunction_read_request_body_deferring_parse_failure
Parse the body, returning a parse failure instead of raising it. A body that fails to parse is still a request from a known caller, so auth m
litellm/proxy/auth/user_api_key_auth.py:2125
↓ 6 callersFunction_ready_key
(proxy: Gateway, session: CliSession, *, subject: str = SUBJECT)
tests/integration/authorization/test_cli_sso_login_ui_disabled.py:257
↓ 6 callersFunction_real_logging_obj
( *, call_type: str = "aresponses", litellm_params: dict[str, object] | None = None, callbacks
tests/unit/responses/test_streaming_iterator.py:532
↓ 6 callersFunction_realtime_http_route_call
( route: str, logging_obj: Logging, custom_llm_provider: str, api_base: str, client: Async
tests/unit/realtime_api/test_main.py:1932
↓ 6 callersFunction_realtime_session_reply
(raw: str, expires_at: int)
tests/integration/providers/test_realtime_webrtc_http.py:60
↓ 6 callersFunction_realtime_websocket_logging
(*failure_callbacks: CustomLogger)
tests/unit/realtime_api/test_main.py:1289
↓ 6 callersFunction_recent_tool_results
Extract the current turn's tool result payloads from the request messages. Tool results are `role == "tool"` messages that sit at the tail of the
litellm/router_strategy/adaptive_router/hooks.py:114
↓ 6 callersMethod_reconstruct_diarized_transcript
Reconstructs a diarized transcript from words with speaker information. Args: words: List of word objects with speaker,
litellm/llms/deepgram/audio_transcription/transformation.py:163
↓ 6 callersMethod_record_error
( prom_logger: Optional["PrometheusLogger"], error_type: str )
enterprise/litellm_enterprise/proxy/common_utils/check_batch_cost.py:464
↓ 6 callersMethod_record_scan_id
Surface the AIRS scan id on the response, so allowed calls are auditable too.
litellm/proxy/guardrails/guardrail_hooks/panw_prisma_airs/panw_prisma_airs.py:827
↓ 6 callersFunction_recorded
(request_data: dict)
tests/unit/enterprise/enterprise_callbacks/test_secret_detection.py:344
↓ 6 callersFunction_recorded_latencies
(cache: DualCache)
tests/unit/router_strategy/test_lowest_latency.py:46
↓ 6 callersMethod_recording_add
()
tests/unit/proxy/management_endpoints/test_ui_sso.py:6808
↓ 6 callersFunction_recording_langfuse
()
tests/unit/proxy/common_utils/test_check_responses_cost.py:2127
↓ 6 callersMethod_records_of
(self, ids: object)
litellm/router_rust/core.py:3318
↓ 6 callersFunction_recover
( monkeypatch: pytest.MonkeyPatch, database_url: str, digests: set[str] | frozenset[str], wind
tests/integration/spend/test_key_metadata_recovery_probe_bounds.py:154
↓ 6 callersMethod_redact_body_for_logging
Return a shallow copy of ``body`` with secret fields redacted. Soniox's create-transcription body can include ``webhook_auth_header_v
litellm/llms/soniox/audio_transcription/handler.py:302
↓ 6 callersFunction_redact_worker_config_for_logging
Mask sensitive fields in the worker config before it enters a log record. `worker_config` reaches `proxy_startup_event` as either the JSON blob
litellm/proxy/proxy_server.py:1258
↓ 6 callersFunction_redeem_assertion_native
(minter, redeemer, cache=None, client_id=REQUESTING_APP, **overrides)
tests/unit/proxy/mcp_server/test_gateway_dcr_flow.py:3328
↓ 6 callersFunction_redis_answering
A Redis double that answers each key from its minute-less prefix and records every MGET.
tests/unit/router_utils/test_routing_read_batch.py:35
↓ 6 callersFunction_reference
(path: Path)
tests/integration/security/test_api_key_secret_reference.py:121
↓ 6 callersMethod_refresh
(self)
litellm/router_rust/surface/model_list.py:220
↓ 6 callersFunction_refresh_token_source_config
()
tests/unit/proxy/mcp_server/outbound_credentials/test_resolver.py:1412
↓ 6 callersFunction_refusal
(verdict: WebsocketFrameModelVerdict)
tests/unit/proxy/test_openai_ws_passthrough_routes.py:413
↓ 6 callersFunction_refused
(rig: _Rig, response: httpx.Response)
tests/integration/observability/test_bedrock_converse_guardrail_rewrites.py:362
↓ 6 callersFunction_refused
(call: Callable[[], object], detail: str)
tests/integration/authorization/test_responses_id_ownership.py:119
↓ 6 callersFunction_refused
(owned: OwnedProxy, path: str, headers: Headers, query: str)
tests/integration/authorization/test_websocket_rejection_logging.py:138
↓ 6 callersFunction_regional
(make_core: MakeCore, **arguments: object)
tests/unit/test_router/core/test_construction_precall_characterization_precall.py:441
↓ 6 callersFunction_register
(proxy: Gateway, owned: Owned)
tests/integration/spend/test_key_metadata_recovery_dropped_connection.py:109
↓ 6 callersFunction_register
( proxy: ProxyClient, resources: ResourceManager, prefix: str, params: LiteLLMParamsBody )
tests/e2e/llm_translation/test_embeddings_endpoint_e2e.py:104
↓ 6 callersFunction_register
(client: A2AClient, resources: ResourceManager, protocol_version: str)
tests/e2e/a2a/test_a2a_agent_e2e.py:52
↓ 6 callersMethod_register
(self, client: PassthroughClient, resources: ResourceManager, prefix: str)
tests/e2e/llm_translation/test_chat_completions_regression_e2e.py:1380
↓ 6 callersFunction_register_anthropic
( proxy: ProxyClient, resources: ResourceManager, sdk: SdkClients, prefix: str, backend: str = ANTHROPIC_B
tests/e2e/llm_translation/test_anthropic_completion_e2e.py:53
↓ 6 callersFunction_register_deployment
( proxy: ProxyClient, resources: ResourceManager, params: LiteLLMParamsBody, prefix: str, )
tests/e2e/llm_translation/test_messages_prompt_caching_and_tool_search_e2e.py:64
↓ 6 callersFunction_register_mode_following_initializer
Registers like the shipped initializers do: construct, then add the instance to litellm's callbacks.
tests/unit/proxy/guardrails/test_guardrail_registry.py:200
↓ 6 callersMethod_register_off_peak_flat_model
(self, model_key: str, off_peak_pricing: dict)
tests/unit/llms/dashscope/test_dashscope_cost_calculator.py:1001
↓ 6 callersMethod_register_openapi_tools
Register tools from an OpenAPI specification for a given server. This creates "virtual" MCP tools from OpenAPI endpoints that are:
litellm/proxy/mcp_server/mcp_server_manager.py:2918
↓ 6 callersFunction_register_routing_test_row
(monkeypatch: pytest.MonkeyPatch, mode: str)
tests/unit/test_router_model_cost_isolation.py:2307
↓ 6 callersFunction_register_tts
(proxy: ProxyClient, resources: ResourceManager)
tests/e2e/llm_translation/test_audio_speech_e2e.py:37
↓ 6 callersFunction_registered_v2_logger
The proxy's registered V2 logger, read without importing the proxy. Request paths call this (the router's ``route`` phase among them), so importi
litellm/integrations/otel/logger.py:1006
↓ 6 callersMethod_registry
()
tests/unit/proxy/policy_engine/test_attachment_registry.py:539
↓ 6 callersFunction_rejecting_native_output
(message: str)
tests/integration/providers/test_bedrock_native_structured_output_wire.py:977
↓ 6 callersMethod_reload_admitted_key
Reload the live key record an admitted envelope references and re-check live policy. Resolving the current ``UserAPIKeyAuth`` (cache first, t
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:1611
↓ 6 callersFunction_remaining_days
(tag: str)
tests/integration/spend/test_daily_tag_spend_retention.py:53
↓ 6 callersFunction_replay
( answers: Sequence[httpx.Response], sent: list[dict[str, object]] )
tests/unit/llms/vertex_ai/gemini/test_rejected_thought_signature_retry.py:88
↓ 6 callersFunction_replay_body
(model: str, marker: str, loads: Sequence[JsonValue], **fields: JsonValue)
tests/integration/authorization/test_tool_search_replay_allowlist.py:24
↓ 6 callersFunction_replayed
The first of up to ``REPLAY_ATTEMPTS`` identical streams served without an upstream call. A streamed request can reach the upstream once per prox
tests/integration/caching/test_response_cache_empty_output.py:148
↓ 6 callersFunction_replayed_reasoning_counts
(upstream: respx.Route)
tests/unit/llms/anthropic/messages/responses_adapters/test_responses_adapters_handler.py:1361
↓ 6 callersFunction_replayed_roles_and_texts
(session: ChatCompletionSession)
tests/unit/responses/litellm_completion_transformation/test_session_handler.py:1102
↓ 6 callersFunction_request
(custom_id: str, text: str)
tests/unit/batches/test_batch_line_items.py:32
↓ 6 callersMethod_request
(self, headers: Mapping[str, str])
tests/unit/proxy/pass_through_endpoints/test_llm_pass_through_endpoints.py:6724
↓ 6 callersMethod_request
(self, content_type: str = "application/json")
tests/unit/proxy/pass_through_endpoints/test_llm_pass_through_endpoints.py:7967
↓ 6 callersMethod_request
( self, *, actor: UserAPIKeyAuth | None = None, metadata_name: str = "metadata
tests/unit/test_router/test_router.py:13176
↓ 6 callersFunction_request_body
(request: Request)
tests/integration/providers/_responses_bridge_prompt_cache_breakpoint.py:249
↓ 6 callersFunction_request_body
(model: str, target: str)
tests/integration/observability/test_panw_airs_block_masked_response.py:182
↓ 6 callersFunction_request_body
(respx_mock)
tests/unit/llms/edenai/chat/test_edenai_chat_transformation.py:79
↓ 6 callersFunction_request_data
(provider, assistant_client, **kwargs)
tests/unit/llms/openai/test_openai.py:2111
↓ 6 callersFunction_request_mock_without_trace_headers
()
tests/unit/proxy/test_litellm_pre_call_utils.py:4663
↓ 6 callersFunction_request_rows
(rows: Sequence[Mapping[str, JsonValue]])
tests/integration/routing/test_complexity_router_databricks_classifier.py:255
↓ 6 callersFunction_request_trace
( *, root_name: str = "POST", root_kind: str = "server", route: str = "/v1/messages", gena
tests/e2e_harness/logging/test_span_selection.py:97
↓ 6 callersFunction_request_with_headers
(headers: dict, *, client_host: str = "127.0.0.1")
tests/unit/proxy/auth/test_oauth2_proxy_hook.py:29
↓ 6 callersFunction_require_all_tags
( deployments: _DeploymentPool, required_set: frozenset[str], )
litellm/router_strategy/tag_based_routing.py:243
↓ 6 callersFunction_require_run
Return the run or raise 404. For non-admin callers, also enforce key ownership.
litellm/proxy/management_endpoints/workflow_management_endpoints.py:200
↓ 6 callersFunction_rescript
(handles: Sequence[ScenarioHandle], response: RoutedResponse)
tests/integration/spend/test_background_response_billing_proxies.py:77
↓ 6 callersFunction_resend_body
(signed: Body, placeholders: Callable[[Body], Body | None])
litellm/llms/vertex_ai/gemini/rejected_thought_signature_retry.py:110
↓ 6 callersFunction_reserve
(valid_token, cost, key_cache, proxy_logging_obj)
tests/unit/proxy/test_budget_reservation.py:91
↓ 6 callersFunction_reset_callbacks_and_settle_pending_logs
()
tests/unit/a2a_protocol/test_cost_calculator.py:16
↓ 6 callersMethod_resolution_case
( self, source: str, caller: str, server_id: str, *, is_byok:
tests/unit/proxy/management_endpoints/test_mcp_management_endpoints.py:9980
↓ 6 callersFunction_resolve
(result: NativeOutputAttemptResult)
tests/unit/llms/bedrock/chat/test_native_output_fallback.py:70
↓ 6 callersFunction_resolve_header
( header_value: str, allowed_team_ids: set[str], fallback_to_db_teams: bool, fake_get_team,
tests/unit/proxy/auth/test_handle_jwt.py:7079
↓ 6 callersMethod_resolve_max_tokens
Extract max_tokens and validate against thinking.budget_tokens. Anthropic API requires ``max_tokens > thinking.budget_tokens``. If th
litellm/integrations/websearch_interception/handler.py:1315
↓ 6 callersFunction_resolve_provider
( *, data: dict, request: Request, )
litellm/proxy/vector_store_files_endpoints/endpoints.py:502
← previousnext →12,201–12,300 of 202,479, ranked by callers