MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 11 callersFunction_start_device_flow
(client: TestClient)
tests/unit/proxy/anthropic_endpoints/test_gateway_endpoints.py:218
↓ 11 callersFunction_storage
(allocated: int, live: int, dead: int, inserted: int | None = None, deleted: int | None = None)
tests/unit/proxy/test_spend_log_cleanup.py:1729
↓ 11 callersFunction_store_rows
(store_id: str)
tests/integration/authorization/test_rag_team_provider_credentials.py:831
↓ 11 callersFunction_stream_chunk
(delta: Delta, finish_reason: str = None, usage: Usage = None)
tests/unit/llms/anthropic/messages/adapters/test_streaming_iterator_tool_args.py:393
↓ 11 callersFunction_stream_finish_reasons
(chunks: list)
tests/unit/llms/vertex_ai/gemini/test_long_decode.py:760
↓ 11 callersMethod_strip_markdown_json
Strip markdown code block wrapper from JSON content if present. SAP GenAI Hub with Anthropic models sometimes returns JSON wrapped in
litellm/llms/sap/chat/transformation.py:426
↓ 11 callersFunction_sync_client
A sync Azure client (passes ``isinstance(.., AzureOpenAI)``).
tests/unit/llms/azure/batches/test_handler.py:75
↓ 11 callersFunction_sync_guardrail_info_to_logging_obj
Copy standard_logging_guardrail_information from request_data into logging_obj. The @log_guardrail_information decorator writes guardrail info to
litellm/integrations/custom_guardrail.py:1885
↓ 11 callersFunction_team
(proxy: Gateway, cleanups: ExitStack, **fields: JsonValue)
tests/integration/management/test_ptu_share_guards.py:114
↓ 11 callersFunction_thinking_deltas
(events: List[dict])
tests/unit/llms/anthropic/messages/adapters/test_streaming_iterator_first_delta.py:303
↓ 11 callersFunction_token_rates_cost
(row: Mapping[str, object], usage: Usage)
tests/unit/test_cost_calculator.py:427
↓ 11 callersFunction_tombstones
(team_id: str)
tests/integration/management/test_team_delete_large_membership.py:124
↓ 11 callersFunction_tool_input
(shape: str, marker_hex: str)
tests/integration/providers/test_vertex_anthropic_structured_output_wire.py:248
↓ 11 callersFunction_tracked_engine
()
tests/unit/proxy/db/test_log_db_metrics.py:16
↓ 11 callersFunction_transaction
( request_id: str, date: str = "2026-07-25", tool_names: tuple = ("tool_a",), spend: float = 1
tests/unit/proxy/db/test_spend_log_tool_index.py:290
↓ 11 callersMethod_transform
(self, url: str, body: dict, model: str = MODEL, target_model: str = "")
tests/unit/llms/bedrock/files/test_bedrock_files_transformation.py:1932
↓ 11 callersMethod_transform
(self, input, model="openai.gpt-5.6-sol")
tests/unit/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py:917
↓ 11 callersMethod_transform_messages_helper
Strip fields not permitted by FireworksAI from messages.
litellm/llms/fireworks_ai/chat/transformation.py:503
↓ 11 callersFunction_transform_response
( config: ScaleDownChatConfig, model: str, payload: dict, request_data: dict | None = None )
tests/unit/llms/scaledown/chat/test_scaledown_chat_transformation.py:33
↓ 11 callersMethod_transform_responses_api_content_to_chat_completion_content
Transform a Responses API content into a Chat Completion content Note: This function should not be called with None content.
litellm/responses/litellm_completion_transformation/transformation.py:2749
↓ 11 callersMethod_transform_system_message
( self, messages: list[AllMessageValues], model: str | None = None, deployment
litellm/llms/bedrock/chat/converse_transformation.py:1722
↓ 11 callersMethod_translate_openai_content_to_anthropic
( self, choices: list[Choices], tool_name_mapping: dict[str, str] | None = None, )
litellm/llms/anthropic/messages/adapters/transformation.py:1551
↓ 11 callersFunction_turn
(*messages: dict)
tests/unit/router_utils/pre_call_checks/test_prompt_caching_deployment_check.py:935
↓ 11 callersFunction_units_upserts
(prisma: MagicMock)
tests/unit/proxy/guardrails/test_usage_tracking.py:56
↓ 11 callersMethod_unmask_response_event
Apply Presidio PII unmasking to backend events before forwarding to the client. Handles two shapes: - ``response.com
litellm/responses/streaming_iterator.py:2949
↓ 11 callersFunction_update_database_and_spend_counters
The reservation is reconciled before the spend is persisted, from its own read. One spend counter batch then spans the database write and the coun
litellm/proxy/hooks/proxy_track_cost_callback.py:1276
↓ 11 callersFunction_upgrades
(gateway: Gateway, scenario_id: str)
tests/integration/providers/test_openai_realtime_transcription_url.py:200
↓ 11 callersFunction_upstream_body
(request: Request)
tests/integration/providers/test_cloudflare_clef_omni_media_wire.py:161
↓ 11 callersFunction_upstream_error_request
()
tests/unit/proxy/pass_through_endpoints/test_pass_through_endpoints.py:5229
↓ 11 callersFunction_upstream_stream
( delta: dict[str, object], text: str = "call 555-0100", tool_uses: tuple[tuple[str, str], ...] = () )
tests/unit/proxy/guardrails/test_anthropic_sse.py:70
↓ 11 callersFunction_upstream_video_id
(video: Video)
tests/integration/compatibility/test_missing_body_param_status.py:1873
↓ 11 callersFunction_user
(text: str)
tests/integration/providers/test_gemini_thinking_replay_wire.py:73
↓ 11 callersFunction_user
()
tests/unit/proxy/video_endpoints/test_endpoints.py:228
↓ 11 callersFunction_validate_team_member_add_permissions
Validate if user has permission to add members to the team. Standard users can self-join an *available team*, but the bypass must not be allo
litellm/proxy/management_endpoints/team_endpoints.py:3180
↓ 11 callersFunction_validate_token_response
Raise HTTPException 403 if any validation rule doesn't match the token response. Supports dot-notation for nested fields (e.g. ``"team.enterprise
litellm/proxy/mcp_server/discoverable_endpoints.py:624
↓ 11 callersFunction_vertex_peer
(status: int = 200, error: str = "scripted vertex rejection")
tests/integration/providers/test_google_genai_vertex_routes_wire.py:143
↓ 11 callersFunction_video_model
(scenario: Scenario, wire: Wire)
tests/integration/compatibility/test_missing_body_param_status.py:1847
↓ 11 callersFunction_video_target
(request: Request)
tests/integration/compatibility/test_missing_body_param_status.py:1864
↓ 11 callersFunction_video_upstream
(seen: list[Request])
tests/integration/compatibility/test_missing_body_param_status.py:1839
↓ 11 callersFunction_wait_until
(condition: Callable[[], bool], timeout_seconds: float = 5.0)
tests/unit/proxy/db/test_pgbouncer.py:350
↓ 11 callersFunction_websearch_logger_cls
()
tests/unit/proxy/proxy_server/test_proxy_config.py:6307
↓ 11 callersFunction_whole_batch
Every line of a batch small enough for one part.
tests/unit/batches/test_batch_line_items.py:85
↓ 11 callersFunction_windowed_router
Real Router; each deployment is (group, provider_model, declared window or None). None means no declared override on a model the cost map does not
tests/unit/router_strategy/test_complexity_router.py:14531
↓ 11 callersFunction_wire_body
(**kwargs: object)
tests/unit/llms/volcengine/test_volcengine_video_transformation.py:839
↓ 11 callersFunction_wire_view_setup
(prisma_client: PrismaClient, probe: AsyncMock)
tests/unit/proxy/utils/prisma_and_spend/test_prisma_client_lifecycle.py:184
↓ 11 callersMethod_write
(self, change: Callable[[list[object]], object])
litellm/router_rust/surface/indexes.py:94
↓ 11 callersFunction_write_config
( directory: Path, upstream_url: str, models: tuple[str, ...], litellm_settings: Mapping[str,
tests/integration/configuration/test_json_logs_startup.py:89
↓ 11 callersFunction_write_token_file
A stored login: `key=None` is the metadata half of a keychain-backed pair, a key is a file-backed one.
tests/unit/proxy/client/cli/test_auth_commands.py:57
↓ 11 callersMethod_zeroed
(model_info=None, litellm_params=None, supplied=None)
tests/unit/proxy/management_endpoints/test_ptu_model_settings.py:786
↓ 11 callersFunctiona
litellm/proxy/_experimental/out/_next/static/chunks/2jxpy6eth1zvy.js:1
↓ 11 callersMethodacancel_kwargs
(self)
tests/unit/proxy/batches_endpoints/test_endpoints.py:2469
↓ 11 callersMethodaclose
(self)
litellm/tracing/exporter.py:283
↓ 11 callersMethodaclose
(self)
tests/unit/mcp_client/test_mcp_client.py:4019
↓ 11 callersMethodaclose
(self)
tests/unit/router_utils/test_fallback_event_handlers.py:1971
↓ 11 callersMethodacreate_sandbox
(self)
tests/unit/litellm_core_utils/test_chat_completion_agentic_loop.py:99
↓ 11 callersMethodadd_agent_to_db
Add an agent to the database. If ``agent_id`` is provided, it is used as the primary key for the new row (otherwise the DB g
litellm/proxy/agent_endpoints/agent_registry.py:581
↓ 11 callersMethodadd_event
(self, event: "EventStreamMessage", headers: Mapping[str, object])
litellm/llms/bedrock/chat/invoke_handler.py:436
↓ 11 callersFunctionadd_provider_specific_params_to_optional_params
Add provider specific params to optional_params
litellm/utils.py:6236
↓ 11 callersMethodadd_subpath_route
Add wildcard route for sub-paths
litellm/proxy/pass_through_endpoints/pass_through_endpoints.py:3937
↓ 11 callersFunctionagent_launch_args
Extra CLI args an agent needs to actually honor the proxy. Claude Code and OpenCode respect the exported env vars, so they get nothing here;
litellm/proxy/client/cli/commands/agents.py:436
↓ 11 callersMethodaiter_bytes
(self)
litellm/passthrough/main.py:184
↓ 11 callersFunctionanthropic_error
(status: int, message: str, *, content_type: str = "application/json")
tests/integration/routing/test_payment_required_mapping.py:64
↓ 11 callersFunctionanthropic_peer
(request: Request)
tests/integration/providers/_cache_control_marks_support.py:334
↓ 11 callersFunctionanthropic_sse_error_events
The ``error`` events of the stream, each as an ``event:`` line and the ``data:`` line assembly read. A stream whose upstream fails after it start
litellm/llms/anthropic/messages/stream_assembly.py:623
↓ 11 callersFunctionappend_query_params
(url: str | None, params: dict)
litellm/proxy/proxy_cli.py:175
↓ 11 callersMethodapply
``apply_changes``: assigned settings, then ``update_settings``, upserts, deletes and the Redis cache, in that order. An unknown name refuses t
litellm/router_rust/core.py:2062
↓ 11 callersMethodapply_authenticated_baggage_tags
( request: Request, request_data: MutableMapping[str, object], # mutable-ok: restore authenti
litellm/proxy/litellm_pre_call_utils.py:2449
↓ 11 callersMethodapply_key_tags_pre_auth
Merge key metadata tags into request_data before _tag_max_budget_check.
litellm/proxy/litellm_pre_call_utils.py:2502
↓ 11 callersMethodapply_pending
( self, apply: Callable[[LedgerTransaction, tuple[DeliveryProjection, ...]], Awaitable[None]],
litellm/repositories/accounting_delivery_repository.py:151
↓ 11 callersFunctionapply_search_free_tier
Zero the grounding queries that landed under the model's shared free allowance, everything else is unchanged.
litellm/proxy/hooks/search_free_tier.py:89
↓ 11 callersFunctionapply_vcr_auto_marker_to_items
Auto-apply ``pytest.mark.vcr`` to collected items. Skip semantics (in priority order): 1. ``vcr_disabled()`` — global env-var off-switch (``
tests/_vcr_conftest_common.py:1396
↓ 11 callersMethodarefresh_model_info
(self, *, client: "AsyncHTTPHandler | None" = None)
litellm/router_rust/core.py:2327
↓ 11 callersMethodartifact_name
(self, base_point: str)
scripts/lint_base_counts.py:63
↓ 11 callersFunctionas
litellm/proxy/_experimental/out/_next/static/chunks/138wymdv6q3wy.js:1
↓ 11 callersFunctionasend_message_streaming
Async: Send a streaming message to an A2A agent. If litellm_params contains custom_llm_provider, routes through the completion bridge.
litellm/a2a_protocol/main.py:658
↓ 11 callersFunctionask
(*, marked: bool)
tests/integration/providers/_cache_control_marks_support.py:90
↓ 11 callersFunctionassert_cut_short
(elapsed: float, *, injected_delay: float)
tests/integration/_support/timing.py:27
↓ 11 callersFunctionassert_daily_window_reported
(rig: BudgetRig, key: str, model: str)
tests/integration/management/test_budget_reset_windows.py:229
↓ 11 callersFunctionassert_recorded_as_a_deployment_failure
(rig: Rig, cell: str, status: str, count: float = 1.0)
tests/integration/observability/test_prometheus_provider_auth_failure.py:482
↓ 11 callersMethodasync_failure_handler
Restores trace_id/session_id contextvars once this attempt's own failure logging (including any nested calls its callbacks trigger) is fully d
litellm/litellm_core_utils/litellm_logging.py:4883
↓ 11 callersMethodasync_get_active_cooldowns
( self, model_ids: Sequence[str], parent_otel_span: Span | None )
litellm/router_utils/cooldown_cache.py:40
↓ 11 callersMethodasync_pre_call_deployment_hook
Persist/update the API-key -> deployment mapping for this request. Why pre-call? - LiteLLM runs async success callbacks via
litellm/router_utils/pre_call_checks/deployment_affinity_check.py:586
↓ 11 callersMethodasync_pre_call_hook
Pre-call hook to scan the request for security threats before sending to LLM. Args: user_api_key_dict: User API key auth
litellm/proxy/guardrails/guardrail_hooks/pillar/pillar.py:324
↓ 11 callersMethodattach_redis_cache
Attach a Redis backend if this DualCache does not already have one. No-op when ``redis_cache`` is None or when Redis was already set
litellm/caching/dual_cache.py:111
↓ 11 callersFunctionauth
(ep: ModelEndpoint)
tests/unit/harness/test_endpoint.py:80
↓ 11 callersFunctionawait_converged
Poll until a read satisfies `converged` or `timeout` elapses. Polls before testing the deadline, so a zero or already-spent budget still gets one
tests/e2e/proxy_client.py:393
↓ 11 callersFunctionb
litellm/proxy/_experimental/out/_next/static/chunks/1f2vcab0yti03.js:53
↓ 11 callersFunctionb
()
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 11 callersMethodbase_validate_azure_environment
( cls, headers: dict[str, str], # mutable-ok: mirrors override contract litellm_param
litellm/llms/azure/common_utils.py:1050
↓ 11 callersMethodbatch_get_cache
(self, key_list, **kwargs)
tests/unit/caching/test_dual_cache.py:549
↓ 11 callersFunctionbedrock_labels
(request: Request)
tests/integration/providers/_cache_control_marks_support.py:378
↓ 11 callersMethodbuild
(e,t,n)
litellm/proxy/_experimental/out/_next/static/chunks/0tx19jeq-wskb.js:2
↓ 11 callersFunctionbuild_agent_skill_archive
( content: bytes, *, skill_path: str | None = None, plugin_name: str | None = None )
litellm/proxy/discovery_endpoints/agent_skills_archive.py:115
↓ 11 callersFunctionbuild_chat_request
( messages: Sequence[AllMessageValues], optional_params: Mapping[str, object], )
litellm/llms/microsoft_365_copilot/chat/transformation.py:205
↓ 11 callersFunctionbuild_conversation
(messages: &[ChatMessage])
litellm-rust/crates/core-utils/src/prompt_templates/factory.rs:72
↓ 11 callersFunctionbuild_skill_archive
Return the installable archive for an upload, or None when it holds no root SKILL.md.
litellm/proxy/discovery_endpoints/agent_skills_archive.py:97
↓ 11 callersMethodbuild_stdio_env
Resolve stdio env values, supporting header-driven placeholders.
litellm/proxy/mcp_server/mcp_server_manager.py:4131
↓ 11 callersMethodbuild_web_search_tool_result_block
Build an Anthropic-native ``web_search_tool_result`` content block. Native Anthropic clients (Claude Desktop, the Anthropic SDK, the
litellm/integrations/websearch_interception/transformation.py:412
↓ 11 callersFunctionbyok_token
Exchange an authorization code for a short-lived BYOK session JWT. 1. Validates the authorization code and PKCE challenge. 2. Stores the
litellm/proxy/mcp_server/byok_oauth_endpoints.py:876
↓ 11 callersFunctioncV
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:10
← previousnext →6,101–6,200 of 202,479, ranked by callers