MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 17 callersFunction_patch_paths
(monkeypatch, tmp_path)
tests/unit/proxy/client/cli/test_up_commands.py:32
↓ 17 callersFunction_plan
(url: str = UPSTREAM, run_as_user: str | None = None)
tests/unit/proxy/db/test_pgbouncer.py:53
↓ 17 callersFunction_policies_sent_to_sts
(callers: tuple[BaseAWSLLM, ...], session_name: str)
tests/unit/llms/bedrock/test_web_identity_session_policy.py:655
↓ 17 callersFunction_prepare_mcp_server_data
Helper function to prepare MCP server data for database operations. Handles JSON field serialization for mcp_info and env fields. Args:
litellm/proxy/mcp_server/db.py:403
↓ 17 callersFunction_priced_shield_guardrail
(**pricing)
tests/unit/proxy/guardrails/guardrail_hooks/azure/test_azure_prompt_shield.py:360
↓ 17 callersMethod_process_hidden_params_and_response_cost
Resolve hidden params, compute response cost, and emit the standard logging payload.
litellm/litellm_core_utils/litellm_logging.py:3120
↓ 17 callersFunction_prompt
(label: str)
tests/integration/providers/test_bedrock_mantle_context_overflow_fallbacks_wire.py:369
↓ 17 callersFunction_provider
(slug, **overrides)
tests/unit/llms/openai_like/test_dynamic_config.py:11
↓ 17 callersFunction_proxy
(prisma)
tests/unit/proxy/management_endpoints/test_access_group_management.py:879
↓ 17 callersFunction_proxy_with_stubbed_reload
delete_access_group finishes by reloading the router and judging what it serves afterwards. Both collaborators it reaches for there are module glo
tests/unit/proxy/management_endpoints/test_access_group_management.py:887
↓ 17 callersFunction_ptu_ceiling_for
(team_id: str, model_group: str, tpm_limit: int, ratio: float, cached_ratio: float = 0.0)
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:8829
↓ 17 callersFunction_quality_group
(strategy="latency-based-routing")
tests/unit/router_strategy/test_router_routing_groups.py:393
↓ 17 callersFunction_query
( gateway: Gateway, key: str, model: str, retrieval_config: Mapping[str, JsonValue], *,
tests/integration/authorization/test_rag_team_provider_credentials.py:783
↓ 17 callersFunction_rates
(replica: Gateway, model: str)
tests/integration/spend/test_background_interaction_settlement.py:353
↓ 17 callersFunction_reject
The single place this adapter raises its own errors.
litellm/llms/scaledown/chat/transformation.py:65
↓ 17 callersFunction_renamed
(attempt: _Attempt, value: int)
tests/integration/providers/test_bedrock_mantle_output_limit_wire.py:291
↓ 17 callersFunction_repository
( database: _FakeDatabase, proxy_reads: _ProxyReads | None = None )
tests/unit/repositories/test_daily_activity_repository.py:218
↓ 17 callersFunction_resolve_allowed_fails_from_policy
Match *exception* against *policy* and return the configured allowed-fail count, or None.
litellm/router_utils/cooldown_handlers.py:180
↓ 17 callersMethod_resolve_job_routing
Resolve (model_id, batch_id) for a managed-object row, where model_id is a router deployment id and batch_id is the raw provider batc
enterprise/litellm_enterprise/proxy/common_utils/check_batch_cost.py:470
↓ 17 callersMethod_resolve_project_name
Resolve the target Phoenix/Arize project for this request. Proxy priority: ``user_api_key_auth_metadata.phoenix_project_name_overrid
litellm/integrations/arize/arize_phoenix.py:265
↓ 17 callersFunction_responses_payload
(model: str)
tests/unit/llms/azure_ai/responses/test_azure_ai_responses_transformation.py:46
↓ 17 callersMethod_router
ids_by_model: model group -> deployment ids the router knows. The fake mirrors the real select(kind="healthy") contract, including how it say
tests/unit/router_strategy/test_complexity_router.py:16090
↓ 17 callersFunction_select_call
The find_many call: (sql, *params). The count call comes first.
tests/unit/proxy/management_endpoints/management_v1/test_budgets.py:108
↓ 17 callersFunction_send
(gateway: Gateway, endpoint: str, *blocks: JsonValue, stream: bool = False)
tests/integration/observability/test_bedrock_guardrail_attachments.py:295
↓ 17 callersFunction_session_updates
(observed: tuple[dict[str, JsonValue], ...])
tests/integration/providers/test_realtime_transcription_guardrail_session_update.py:759
↓ 17 callersFunction_setup_unscoped_list_files_route_over_real_hook
Wire GET /v1/files to a real managed-files hook over `rows`, with no provider credentials in the process. The neighbouring setup stubs the hook wi
tests/unit/proxy/openai_files_endpoint/test_files_endpoint.py:5048
↓ 17 callersFunction_shadow_over_unpriced_target
(rig: AliasRig, scenario: Scenario)
tests/integration/authorization/test_hidden_alias_budget_bypass.py:585
↓ 17 callersFunction_single
(payload: dict)
tests/unit/proxy/management_endpoints/test_mcp_connector_import.py:13
↓ 17 callersFunction_spend_reader
()
tests/unit/proxy/roi_calculator/test_sync.py:206
↓ 17 callersFunction_sse
(*chunks: Mapping[str, object])
tests/unit/llms/vertex_ai/gemini/test_long_decode.py:118
↓ 17 callersFunction_stream
( gateway: Gateway, endpoint: Endpoint, body: Mapping[str, JsonValue], *, key: str | None = None )
tests/integration/providers/test_bedrock_converse_stream_event_frames_wire.py:224
↓ 17 callersFunction_success
(access_token: str = "m2m-token", **extra: object)
tests/unit/proxy/mcp_server/outbound_credentials/test_client_credentials.py:48
↓ 17 callersFunction_toolset
(scenario: Scenario, server_id: str, tool: str)
tests/integration/mcp/test_mcp_toolsets.py:44
↓ 17 callersFunction_transport
(upstream: Callable[[httpx.Request], httpx.Response])
tests/unit/proxy/hooks/test_autorouter_baseline_cache.py:158
↓ 17 callersFunction_user_content_on_wire
( bridge: _Bridge, marker: str, *, model: str = "gpt-6.1-sol", stream: bool = False )
tests/integration/providers/test_responses_bridge_input_audio_wire.py:120
↓ 17 callersFunction_verdict_response
Build a mocked Repello analyze response with the given verdict.
tests/unit/proxy/guardrails/guardrail_hooks/test_repelloai.py:32
↓ 17 callersFunction_vertex_deployment
(gateway: Gateway, scenario: Scenario, api_base: str)
tests/integration/providers/test_vertex_anthropic_inline_tools_beta_wire.py:126
↓ 17 callersFunction_video
(video_id: str, hidden_params: dict[str, object] | None = None)
tests/unit/proxy/video_endpoints/test_utils.py:219
↓ 17 callersFunction_wire_member_delete_tx
/team/member_delete's four cleanups and /team/member_update's role rewrite, plus the advisory-lock re-read that guards them, run inside one transa
tests/unit/proxy/management_endpoints/test_team_endpoints.py:199
↓ 17 callersMethod_zero_cost_warnings
(caplog: pytest.LogCaptureFixture)
tests/unit/litellm_core_utils/test_litellm_logging.py:767
↓ 17 callersFunctiona
()
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
↓ 17 callersMethodabort
()
litellm/proxy/_experimental/out/_next/static/chunks/1f2vcab0yti03.js:5
↓ 17 callersMethodacreate_kwargs
Exact kwargs forwarded to litellm.acreate_batch.
tests/unit/proxy/batches_endpoints/test_endpoints.py:167
↓ 17 callersMethodacreate_sandbox
(self, **kwargs)
tests/unit/integrations/code_interpreter_interception/test_handler.py:55
↓ 17 callersFunctionactiveTierRows
(value: ActiveTierSet)
ui/litellm-dashboard/src/components/add_model/tier_rows.ts:65
↓ 17 callersMethodadd
(self, value: object)
litellm/router_rust/surface/indexes.py:216
↓ 17 callersFunctionadd_team_callbacks
Add a success/failure callback to a team Use this if if you want different teams to have different success/failure callbacks Parameters
litellm/proxy/management_endpoints/team_callback_endpoints.py:275
↓ 17 callersMethodaextraction
( self, model: str, custom_llm_provider: str, logging_obj: LiteLLMLoggingObj |
litellm/llms/custom_httpx/llm_http_handler.py:1954
↓ 17 callersFunctionaiter_sse_lines
(text_chunks: AsyncIterable[str])
litellm/llms/custom_httpx/sse_lines.py:68
↓ 17 callersFunctionalias_values
The key aliases holding a series of their own on the metric; the shared overflow series is not one.
tests/integration/_support/prometheus_series.py:248
↓ 17 callersMethodallows
(self, group: TelemetryGroup)
litellm/telemetry/consent.py:45
↓ 17 callersMethodamoderation
(self, *args: object, **kwargs: object)
litellm/router_rust/core.py:2609
↓ 17 callersMethodappend
( self, record: BaselineAccountingRecord )
litellm/proxy/db/baseline_accounting.py:439
↓ 17 callersFunctionassert_loop_stayed_free
(took: float, lags: tuple[float, ...])
tests/unit/litellm_core_utils/event_loop_lag.py:44
↓ 17 callersFunctionassert_once
(received: Sequence[str], markers: Sequence[str])
tests/integration/routing/test_priority_scheduler_queue_cleanup.py:167
↓ 17 callersFunctionassert_one_id
(endpoint: Endpoint, stream: bool, observed: Seen, response_id: str)
tests/integration/providers/_gemini_long_decode_support.py:548
↓ 17 callersFunctionassert_upload_limited
(observed: Timed, day_ends: float, limit: int, holder: str, source: str)
tests/integration/management/_batch_file_caps.py:272
↓ 17 callersMethodasync_get_available_deployment
( self, model: str, messages: list[dict[str, str]] | None = None, input: str |
litellm/router_strategy/lar1_routing.py:93
↓ 17 callersMethodasync_get_healthy_deployments
( self, model: str, request_kwargs: dict, # mutable-ok: the old parameter type, uncha
litellm/router_deprecated.py:714
↓ 17 callersMethodasync_post_call_success_hook
( self, data: dict, user_api_key_dict: UserAPIKeyAuth, response: Any, )
tests/unit/proxy/hooks/test_image_generation_guardrails.py:208
↓ 17 callersMethodasync_pre_call_deployment_hook
( self, kwargs: dict[str, object], call_type: CallTypes | None )
litellm/integrations/code_interpreter_interception/handler.py:303
↓ 17 callersMethodasync_streaming_data_generator
Shared streaming data generator: runs proxy iterator hook, per-chunk hook, cost injection, then yields chunks via serialize_chunk; on
litellm/proxy/common_request_processing.py:4880
↓ 17 callersFunctionattribute_of
(value: object, name: str, default: object = None)
litellm/proxy/common_utils/openai_error_payload.py:73
↓ 17 callersFunctionbackground_streaming_task
Background task to stream response and update cache Follows OpenAI Response Streaming format: https://platform.openai.com/docs/api-refer
litellm/proxy/response_polling/background_streaming.py:132
↓ 17 callersFunctionbad_request
(model: str = "gpt-4o", message: str = "bad input")
tests/unit/test_router/core/scripted_host.py:125
↓ 17 callersMethodbatch
(self)
tests/unit/proxy/batches_endpoints/test_litellm_executed_batches.py:113
↓ 17 callersMethodbuild_container_id
Build a managed container ID with provider and model info encoded. Format: cntr_{base64("litellm:custom_llm_provider:{provider};model_id:{mod
litellm/responses/utils.py:836
↓ 17 callersFunctionbuild_langfuse_client
The REST client for prompt management, ``auth_check`` and the Slack project link. Missing keys are passed through as absent credentials: the serv
litellm/integrations/langfuse/langfuse_sdk.py:1191
↓ 17 callersMethodcalls
(self)
tests/integration/providers/test_strands_decider_agentcore_wire.py:228
↓ 17 callersFunctioncan_access_resource
Return True iff the caller may read/modify a managed resource. The resource's ``created_by`` and ``team_id`` fields must be non-None to match
litellm/llms/base_llm/managed_resources/isolation.py:110
↓ 17 callersFunctionchanged_team_fields
Logical field names whose stored value the request would change. Request and stored row are compared as JSON values so both sides share one repre
litellm/proxy/management_endpoints/team_admin_field_permissions.py:379
↓ 17 callersFunctionchat
(gateway: Gateway, key: str, group: str, content: str, **extra: JsonValue)
tests/integration/routing/test_per_model_rate_limit_fallbacks.py:225
↓ 17 callersMethodchat
( self, key: str, model: str, content: str, *, max_tokens: int
tests/e2e/quota_management/budgets/budget_client.py:284
↓ 17 callersMethodcheck_passthrough_route_access
Check if route is a passthrough route. Supports both exact match and prefix match.
litellm/proxy/auth/route_checks.py:913
↓ 17 callersMethodchunk_parser
(self, chunk: Any)
tests/unit/llms/test_custom_llm.py:46
↓ 17 callersFunctionclassify_prompt
Classify a single user prompt. Falls back to GENERAL when no rule matches. Empty/whitespace-only also returns GENERAL.
litellm/router_strategy/adaptive_router/classifier.py:123
↓ 17 callersFunctioncomplete_sync
(core: RouterCore, model: str = "a", **request: object)
tests/unit/test_router/core/fallbacks_helpers.py:31
↓ 17 callersFunctioncompletion
( model: str, messages: list, api_base: str, model_response: ModelResponse, print_verbose:
litellm/llms/replicate/chat/handler.py:129
↓ 17 callersMethodconfigure
(self, credential=StaticToken("sk-virtual-key"), model=StartOn("claude-auto"), **kwargs)
tests/unit/proxy/client/cli/test_claude_settings.py:417
↓ 17 callersFunctioncooled_down
(core: RouterCore, ids: list[str])
tests/unit/test_router/core/test_retries_cooldowns.py:78
↓ 17 callersMethodcreate
(self, data: Mapping[str, object], include: Mapping[str, object] | None = None)
litellm/repositories/prisma_protocols.py:53
↓ 17 callersFunctioncreate_a2a_client
Create an A2A client for the given agent URL. This resolves the agent card and returns a ready-to-use A2A client. The client can be reus
litellm/a2a_protocol/main.py:836
↓ 17 callersFunctiond
(e,s)
litellm/proxy/_experimental/out/_next/static/chunks/2k-fkubkt19cc.js:26
↓ 17 callersMethoddelete_cache
(self, key: str)
litellm/proxy/auth/login_throttle.py:78
↓ 17 callersMethoddelete_cache
(self, key: str)
tests/unit/proxy/auth/test_auth_checks.py:2382
↓ 17 callersFunctiondeployment_model_info
(litellm_params: Mapping[str, object] | None)
litellm/llms/bedrock/common_utils.py:319
↓ 17 callersMethoddrain
(self)
litellm/proxy/db/request_error_tracking.py:113
↓ 17 callersMethodduring_call_hook
( self, data: dict, user_api_key_dict: UserAPIKeyAuth | None, call_type: CallT
litellm/proxy/utils.py:3803
↓ 17 callersFunctioneffective_skip_tool_message_for_guardrail
(guardrail_to_apply: object)
litellm/llms/base_llm/guardrail_translation/utils.py:239
↓ 17 callersFunctionel
()
litellm/proxy/_experimental/out/_next/static/chunks/37cr8te-omsl0.js:1
↓ 17 callersMethodembedding
( # pyright: ignore[reportIncompatibleMethodOverride] # see acompletion self, model: str, input:
litellm/router.py:272
↓ 17 callersMethodentry
(self)
tests/integration/caching/test_response_cache_chaos.py:85
↓ 17 callersFunctionerror
(status: int, message: str, code: str)
tests/integration/_support/responses_vendor.py:44
↓ 17 callersMethodexecute
( &self, call: ResponsesCall, interceptors: &impl Interceptors<Error>, options
litellm-rust/crates/inference-responses/src/lib.rs:67
↓ 17 callersMethodexists
Check if a record exists.
litellm/repositories/base_repository.py:125
↓ 17 callersFunctionfilter_web_search_deployments
If the request is websearch, filter out deployments that don't support web search
litellm/litellm_core_utils/routing_common_utils.py:452
↓ 17 callersMethodfind_first
(self, where: Mapping[str, object])
enterprise/litellm_enterprise/proxy/hooks/managed_files.py:328
↓ 17 callersMethodfind_many
(self, *, where: Mapping[str, object])
litellm/proxy/management_helpers/utils.py:92
↓ 17 callersMethodfreeze
()
litellm/proxy/_experimental/out/_next/static/chunks/14v3irtjifjt2.js:1
↓ 17 callersFunctionfrom_json
The Python value `json.loads` returns for a JSON document.
litellm-rust/crates/python-compat/src/json.rs:26
← previousnext →3,401–3,500 of 202,479, ranked by callers