MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 8 callersFunction_stream_body
(marker: str)
tests/integration/observability/test_bedrock_converse_stream_post_call_guardrail.py:438
↓ 8 callersFunction_stream_chunk
(delta, finish_reason=None, index=0, model="openai.gpt-oss-20b-1:0")
tests/unit/llms/bedrock/chat/chat_completions/test_bedrock_chat_completions_transformation.py:990
↓ 8 callersFunction_stream_mcp_asgi_response
Call an ASGI MCP handler and return a StreamingResponse so SSE/streaming works. asyncio.create_task copies the current context, so any Conte
litellm/proxy/proxy_server.py:22603
↓ 8 callersFunction_streamed
(gateway: Gateway, endpoint: Endpoint, model: str, **extra: JsonValue)
tests/integration/providers/test_hosted_vllm_stream_stall_wire.py:144
↓ 8 callersFunction_streaming_setup
( g, chunks, cisco_response=None, upstream=None, request_data=None, post_mock=None, )
tests/unit/proxy/guardrails/guardrail_hooks/_cisco_ai_defense_test_utils.py:284
↓ 8 callersFunction_strip_default_port
Return ``netloc`` lowercased with the scheme's default port stripped. ``Llm.Example.com:443`` with scheme ``https`` becomes ``llm.example.com`
litellm/proxy/mcp_server/oauth_utils.py:363
↓ 8 callersFunction_strip_reminder_blocks
Remove every complete reminder block from text, keeping everything written around them. Blocks from different pairs can nest or overlap, which th
litellm/router_strategy/complexity_router/complexity_router.py:564
↓ 8 callersMethod_support
(self, model, llm_router=None, team_id=None)
tests/unit/proxy/test_common_request_processing.py:9944
↓ 8 callersFunction_sync_client
( seen: list[httpx.Request], reply: Callable[[httpx.Request], httpx.Response] = _edited )
tests/unit/llms/dashscope/test_dashscope_image_edit_transformation.py:42
↓ 8 callersMethod_sync_in_memory_spend_with_redis
Ensures in-memory cache is updated with latest Redis values for all provider spends. Why Do we need this? - Optimization to
litellm/router_strategy/budget_limiter.py:734
↓ 8 callersFunction_team
(member_permissions: Sequence[str])
tests/unit/proxy/management_endpoints/test_key_budget_change.py:204
↓ 8 callersFunction_team_info
(gateway: Gateway, team_id: str)
tests/integration/management/test_team_member_lifecycle.py:143
↓ 8 callersMethod_team_member
(self, permissions: list)
tests/unit/proxy/vector_store_endpoints/test_vector_store_endpoints.py:2872
↓ 8 callersFunction_team_member_limiter
( spend_by_model: Mapping[str, float] | None = None, )
tests/unit/proxy/auth/test_fallback_budget.py:62
↓ 8 callersFunction_team_member_model_budget_load_inputs
()
tests/unit/proxy/auth/test_user_api_key_auth_request_flow.py:6132
↓ 8 callersMethod_team_request
(cls, team_id: str | None = "team-a", tags: list[str] | None = None)
tests/unit/test_router/test_router.py:10330
↓ 8 callersFunction_teams_by_alias
An alias lookup over `aliases` (alias -> team_id) that 404s like the real one otherwise.
tests/unit/proxy/auth/test_handle_jwt.py:5912
↓ 8 callersFunction_test_connection_probe
( deployment: Mapping[str, object], )
tests/unit/proxy/health_endpoints/test_health_endpoints.py:709
↓ 8 callersFunction_text
()
tests/integration/sdk/test_ollama_embedding_wire.py:87
↓ 8 callersFunction_text
(text: str)
tests/unit/router_utils/pre_call_checks/test_prompt_caching_deployment_check.py:939
↓ 8 callersFunction_texts
(message: dict)
tests/unit/litellm_core_utils/prompt_templates/test_litellm_core_utils_prompt_templates_mid_conversation_system.py:30
↓ 8 callersFunction_texts
(message: dict)
tests/unit/llms/anthropic/chat/test_anthropic_chat_transformation.py:7401
↓ 8 callersFunction_tiers
(entry)
tests/unit/proxy/common_utils/test_codex_model_catalog.py:87
↓ 8 callersMethod_timeout_ms
The per-statement bound in milliseconds: the batch timeout, or whatever is left of the run budget, whichever is smaller. Cla
litellm/proxy/db/db_transaction_queue/spend_log_cleanup.py:710
↓ 8 callersFunction_tool_call
(arguments, name="f", tc_id="1")
tests/unit/proxy/guardrails/guardrail_hooks/test_vigil_guard.py:622
↓ 8 callersFunction_tool_frame
(identity: str, entries: Sequence[JsonValue], *, first: bool)
tests/integration/messages_endpoint/chat_bridge/test_chat_bridge_multi_tool_call_blocks_wire.py:196
↓ 8 callersFunction_tool_items
(payload: Mapping[str, JsonValue])
tests/integration/providers/test_responses_bridge_tool_search_wire.py:406
↓ 8 callersFunction_tool_request_inputs
(tool_names: list)
tests/unit/proxy/guardrails/guardrail_hooks/test_tool_policy_guardrail.py:29
↓ 8 callersFunction_toolset_rpc
( gateway: Gateway, headers: dict[str, str], name: str, method: str, params: dict[str, object] )
tests/integration/mcp/test_mcp_toolsets.py:49
↓ 8 callersFunction_transaction
( session_id: str = "s1", at: datetime = datetime(2026, 8, 1, 12, 0, 0), tier: str | None = "mediu
tests/unit/proxy/db/test_autorouter_session_rollup.py:312
↓ 8 callersFunction_transform
(payload: object, status_code: int = 200)
tests/unit/llms/dashscope/test_dashscope_rerank_transformation.py:371
↓ 8 callersFunction_transform
(payload: object)
tests/unit/llms/firecrawl/search/test_firecrawl_search_transformation.py:9
↓ 8 callersFunction_transform
(payload: object, query: str = "q=litellm")
tests/unit/llms/brave/search/test_brave_search_transformation.py:9
↓ 8 callersMethod_transform
(self, optional_params, config=None, model="gpt-5.6-sol")
tests/unit/llms/openai/chat/test_openai_gpt_transformation.py:1169
↓ 8 callersMethod_transform_input_file_item_to_file_item
Transform a Responses API input_file item to a Chat Completion file item Args: item: Dictionary containing input_file ty
litellm/responses/litellm_completion_transformation/transformation.py:2727
↓ 8 callersFunction_transform_perplexity_response
(body: object)
tests/unit/llms/perplexity/chat/test_perplexity_chat_transformation.py:778
↓ 8 callersFunction_transform_with_prompt_cache_key
( litellm_params: dict[str, object], prompt_cache_key: object = "session-1" )
tests/unit/llms/mistral/test_mistral_chat_transformation.py:1846
↓ 8 callersMethod_translate
(self, tool_result_content)
tests/unit/llms/anthropic/messages/responses_adapters/test_responses_adapters_transformation.py:1799
↓ 8 callersMethod_translate_streaming_openai_chunk_to_anthropic
( self, choices: list[OpenAIStreamingChoice | StreamingChoices] )
litellm/llms/anthropic/messages/adapters/transformation.py:1908
↓ 8 callersFunction_try_websearch_short_circuit
Attempt to short-circuit a web-search-only request. Claude Code sends web search as a separate, standalone /v1/messages request. For pro
litellm/llms/anthropic/messages/native/handler.py:387
↓ 8 callersFunction_turn_key
(kind: str, minute: str = "12-34")
tests/unit/test_router/test_io_token_rate_limits.py:1648
↓ 8 callersFunction_update_internal_new_user_params
(data_json: dict, data: NewUserRequest)
litellm/proxy/management_endpoints/internal_user_endpoints.py:230
↓ 8 callersFunction_update_litellm_setting
Common utility function to update `litellm_settings` in both memory and config. Args: settings: The settings object to update
litellm/proxy/ui_crud_endpoints/proxy_setting_endpoints.py:1073
↓ 8 callersMethod_update_model_access_group_db
Update spend for every model access group this request is billed against. Args: response_cost: Cost of the request, char
litellm/proxy/db/db_spend_update_writer.py:1678
↓ 8 callersFunction_upstream
(provider: str, model: str)
tests/unit/llms/openai/test_openai.py:790
↓ 8 callersFunction_upstream_url
(gateway: Gateway)
tests/integration/providers/test_vertex_realtime_multi_region_host.py:111
↓ 8 callersFunction_usage
(test_client: TestClient, start_date: str, end_date: str, **headers: str)
tests/unit/proxy/management_endpoints/test_subscription_accounts.py:126
↓ 8 callersMethod_user
(self, role="internal_user", user_id="user-a")
tests/unit/proxy/management_helpers/test_team_member_permission_checks.py:264
↓ 8 callersFunction_user_table
( prisma_client: "PrismaClient | None", )
litellm/proxy/management_endpoints/internal_user_endpoints.py:153
↓ 8 callersFunction_user_teams
(user_id: str)
tests/integration/management/test_team_delete_large_membership.py:96
↓ 8 callersFunction_user_turn
(text: str)
tests/e2e/llm_translation/test_messages_e2e.py:109
↓ 8 callersFunction_validate_member_user_id_provisioning
Restrict adding a caller-chosen user_id that has no user row yet to proxy admins. Team and org admins keep the ability to add users that already
litellm/proxy/management_endpoints/team_endpoints.py:3532
↓ 8 callersFunction_vertex_gemini_response
(text: str = "hello")
tests/unit/llms/vertex_ai/test_vertex.py:1652
↓ 8 callersFunction_vertex_request
(optional_params: dict[str, object], headers: dict[str, str], litellm_params: dict[str, object] | None = None)
tests/unit/llms/vertex_ai/vertex_ai_partner_models/anthropic/test_vertex_ai_partner_models_anthropic_transformation.py:1225
↓ 8 callersFunction_vllm_credential
(api_key: str)
tests/unit/test_utils.py:7629
↓ 8 callersFunction_voice_frames
(*updates: dict[str, JsonValue])
tests/integration/providers/test_realtime_transcription_guardrail_session_update.py:700
↓ 8 callersFunction_wait_for_retirements
(wrapper: PrismaWrapper)
tests/unit/proxy/db/test_prisma_planned_engine_restart.py:169
↓ 8 callersFunction_wait_until_every_worker_serves
(gateway: Gateway, model: str)
tests/integration/management/test_vector_store_file_list_managed_ids.py:350
↓ 8 callersFunction_wait_until_served
(gateway: Gateway, model: str)
tests/integration/management/test_fallback_management_public_names.py:142
↓ 8 callersFunction_wav
(seconds: float = 1.0)
tests/unit/llms/nvidia_riva/audio_transcription/test_handler.py:166
↓ 8 callersFunction_wire_and_result
(guardrail: HeadroomGuardrail, messages: list, returned: list | None = None)
tests/unit/proxy/guardrails/guardrail_hooks/test_headroom.py:2561
↓ 8 callersMethod_wire_body
(**kwargs)
tests/unit/llms/minimax/videos/test_minimax_video_transformation.py:1016
↓ 8 callersMethod_wire_request
(**kwargs)
tests/unit/llms/dashscope/test_dashscope_video_transformation.py:1328
↓ 8 callersFunction_worker_report
Stand-in for a pytest TestReport delivered to the xdist controller. Only the attributes ``aggregate_report_outcome`` reads (``nodeid``, ``whe
tests/unit/test_vcr_classification.py:378
↓ 8 callersMethod_wrap_streaming_iterator_with_enrichment
( callback: object, response: AsyncIterable[_T], hook: _StreamIteratorHook[_T],
litellm/proxy/utils.py:3618
↓ 8 callersFunction_yaml_escape
Escape a string for safe embedding in a double-quoted YAML scalar.
litellm/setup_wizard.py:190
↓ 8 callersMethodabatch_completion_fastest_response
( self, model: str, messages: object, stream: bool = False, **kwargs:
litellm/router_rust/invoke/endpoints.py:2068
↓ 8 callersFunctionabbreviate_api_key
(api_key: str)
litellm/proxy/auth/auth_utils.py:2533
↓ 8 callersFunctionaccess_group_skills
Resolve ``access_skill_names`` of the groups attached to the key and to its team. Key-side groups count only when the group lists the key or its
litellm/proxy/anthropic_endpoints/claude_code_endpoints/claude_code_skill_access.py:116
↓ 8 callersFunctionaccess_windows_config_error
(model_info: Mapping[str, object], *, model_name: str)
litellm/litellm_core_utils/access_windows.py:23
↓ 8 callersMethodaclose
(self)
tests/unit/proxy/test_common_request_processing.py:12137
↓ 8 callersFunctionaclose_if_supported
(stream: object)
litellm/llms/anthropic/messages/native/streaming_iterator.py:357
↓ 8 callersMethodacompletion
(self, *args, **kwargs)
litellm/proxy/example_config_yaml/custom_handler.py:16
↓ 8 callersMethodacompletion
(self, *args, **kwargs)
litellm/completion_extras/litellm_responses_transformation/handler.py:281
↓ 8 callersMethodacquire
(self, bucket: RateBucket)
tests/e2e/cli_harness/rate_limiter.py:86
↓ 8 callersMethodacquire_lock
(self, cronjob_id: str)
tests/unit/proxy/db/test_gateway_request_tracking.py:504
↓ 8 callersFunctionadapt_messages_to_cohere_standard
Build a Cohere ``chatHistory`` list from an OpenAI-format message array. - All messages except the *last user message* are included. The caller p
litellm/llms/oci/chat/cohere.py:106
↓ 8 callersFunctionadd_langfuse_trace_id_to_alert
Returns langfuse trace url - check: -> existing_trace_id -> trace_id -> litellm_call_id
litellm/integrations/SlackAlerting/utils.py:60
↓ 8 callersMethodadd_litellm_failure_callback
Add a failure callback to `litellm.failure_callback`. Auto-routes async callbacks to litellm._async_failure_callback.
litellm/litellm_core_utils/logging_callback_manager.py:95
↓ 8 callersFunctionadd_path_to_api_base
Adds an ending path to an API base URL while preventing duplicate path segments. Args: api_base: Base URL string ending_path
litellm/utils.py:11851
↓ 8 callersMethodadd_team_member
(self, team_id: str, user_id: str, *, max_budget_in_team: float | None = None)
tests/e2e/quota_management/budgets/budget_client.py:483
↓ 8 callersMethodadd_team_member
(self, team_id: str, user_id: str)
tests/e2e/management/management_client.py:421
↓ 8 callersFunctionadd_team_org_context_to_request_body
Return a copy of request_body with organization_id resolved from the target team when the route identifies the team by team_id and the caller
litellm/proxy/auth/auth_checks_organization.py:186
↓ 8 callersMethodadelete_sandbox
(self, *, container, **kwargs)
tests/unit/integrations/code_interpreter_interception/test_handler.py:63
↓ 8 callersFunctionadls_safe_file_name
A Responses API id is base64 behind `resp_`, and the Data Lake service rejects its `=` padding and `/`, so that name drops the padding and maps `/
litellm/integrations/azure_storage/azure_storage.py:40
↓ 8 callersMethodaggregate_queue_updates
Concatenate all updates in the queue to reduce the size of in-memory queue
litellm/proxy/db/db_transaction_queue/spend_update_queue.py:54
↓ 8 callersMethodaiter_bytes
(self)
tests/unit/responses/test_responses_streaming_iterator.py:538
↓ 8 callersMethodalias_of
The group `model_group_alias` rewrites `name` to.
litellm-rust/crates/router-core/src/registry.rs:994
↓ 8 callersMethodallowed_fails_for
Allowed fails for this exception under allowed_fails_policy (was ``get_allowed_fails_from_policy``).
litellm/router_api.py:1285
↓ 8 callersMethodallowsMarkType
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 8 callersFunctionanthropic_error_sse_frame
(status_code: int, raw_message: str)
litellm/anthropic_interface/exceptions/exception_mapping_utils.py:203
↓ 8 callersFunctionanthropic_provider
(chunks: tuple[bytes, ...], *, pause_between_chunks: float = 0)
tests/integration/observability/test_presidio_streaming_output.py:356
↓ 8 callersFunctionapi_key_bucket
(api_key: str | None)
litellm/repositories/daily_activity_buckets.py:19
↓ 8 callersMethodapplied_edits_for_response
``applied_edits`` to attach on the client-visible response. ``compact_20260112`` is included when a new compaction block was synthesi
litellm/llms/anthropic/messages/context_management/result.py:28
↓ 8 callersMethodapply_guardrail
( self, inputs: GenericGuardrailAPIInputs, request_data: Mapping[str, object],
litellm/proxy/guardrails/guardrail_hooks/ovalix/ovalix.py:339
↓ 8 callersMethodapply_replica_identity_full_if_requested
Re-assert REPLICA IDENTITY FULL on LiteLLM's tables when the operator opted in via LITELLM_SET_REPLICA_IDENTITY_FULL. Prisma
litellm-proxy-extras/litellm_proxy_extras/utils.py:1368
↓ 8 callersMethodarc
(e,t,r,n,i,a)
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 8 callersFunctionaresolve_copilot_user_session
( user_id: str, github_token: str, credential_name: str, endpoints: GithubHostEndpoints, )
litellm/llms/github_copilot/per_user_auth.py:389
↓ 8 callersFunctionarize_preset
( *, config_overrides: OpenTelemetryV2Config | None = None, allow_missing_credentials: bool = Fals
litellm/integrations/otel/presets/arize.py:31
↓ 8 callersFunctionassert_completion
(body: Mapping[str, JsonValue])
tests/integration/translation/chat_completions/cross_model/contracts.py:286
← previousnext →9,001–9,100 of 202,479, ranked by callers