MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 8 callersFunction_assert_upstream_headers_recorded
(response)
tests/unit/llms/custom_httpx/test_llm_http_handler.py:5829
↓ 8 callersFunction_async_openai_client
(gateway: Gateway)
tests/integration/providers/test_ollama_prompt_tools_wire.py:194
↓ 8 callersFunction_async_pre_call
(check: ModelRateLimitingCheck, deployment: dict[str, object])
tests/unit/test_router/test_enforce_model_rate_limits.py:490
↓ 8 callersFunction_auth_headers
()
tests/integration/test_oci_proxy_integration.py:174
↓ 8 callersFunction_authorize_post_with_cookie
(client, cookie_jwt: str, api_key: str = "upstream-key")
tests/unit/proxy/mcp_server/test_byok_oauth_endpoints.py:891
↓ 8 callersFunction_authorize_then_callback
Run /authorize for ``server``, then feed the resulting flow back through /callback with the RFC 9207 ``iss`` the authorization server supposedly r
tests/unit/proxy/mcp_server/test_discoverable_endpoints.py:4873
↓ 8 callersFunction_auto_register_kwargs
(prisma_client, user_api_key_cache, jwt_handler, **over)
tests/unit/proxy/auth/test_user_api_key_auth_request_flow.py:2296
↓ 8 callersFunction_auto_register_patches
(*, plaintext_key: str | None = "sk-minted-plaintext")
tests/unit/proxy/auth/test_user_api_key_auth_request_flow.py:2265
↓ 8 callersFunction_awaiting_retry
(logger, queue_attr)
tests/unit/integrations/test_azure_sentinel.py:501
↓ 8 callersFunction_batch_cost_payload
()
tests/unit/proxy/db/test_db_spend_update_writer.py:4425
↓ 8 callersFunction_batch_resolve_access_group_resources
Batch-fetch access groups in a single DB query and return them keyed by access_group_id. Missing/invalid groups are silently omitted.
litellm/proxy/management_endpoints/team_endpoints.py:6028
↓ 8 callersFunction_bedrock_output
(invoke: bool, text: str = "partial output")
tests/integration/observability/test_bedrock_error_request_id.py:108
↓ 8 callersFunction_bedrock_peer
Turn 1 answers with a redacted reasoning block, the way Bedrock does for a safety-redacted Claude turn. Turn 2, the one replaying that history, ge
tests/integration/providers/test_bedrock_converse_redacted_thinking_wire.py:162
↓ 8 callersMethod_bedrock_tier_entry
( model_name: str = "bedrock-tier", aws_region_name: str | None = None, model: str = "
tests/unit/router_strategy/test_complexity_router.py:5356
↓ 8 callersFunction_benchmark_totals
(row: _SessionAggRow)
litellm/proxy/management_endpoints/auto_router_endpoints.py:759
↓ 8 callersFunction_betas
(headers)
tests/unit/llms/anthropic/messages/native/test_anthropic_messages_per_turn_control.py:32
↓ 8 callersFunction_bill_finished_background_response
( response: object, response_id: str, user_api_key_dict: UserAPIKeyAuth, fastapi_response: Res
litellm/proxy/response_api_endpoints/endpoints.py:982
↓ 8 callersFunction_billed_key
(response: httpx.Response)
tests/integration/authorization/test_jwt_auto_register_map_existing_key.py:111
↓ 8 callersMethod_blocked_and_rewritten
( self, rules: list[dict[str, str]], request: dict[str, object], call_type: CallTypesLiteral )
tests/unit/proxy/guardrails/guardrail_hooks/test_tool_permission.py:2406
↓ 8 callersFunction_body
( endpoint: Endpoint, model: str, tools: Sequence[Mapping[str, JsonValue]], *, stream: boo
tests/integration/providers/test_bedrock_converse_lookaround_regex_wire.py:181
↓ 8 callersFunction_body
(messages: tuple[Message, ...])
tests/unit/proxy/middleware/test_gzip_middleware.py:130
↓ 8 callersFunction_bridge_input
( model: str, drop_params: bool | None, messages: Sequence[Mapping[str, object]] | None = None,
tests/unit/completion_extras/litellm_responses_transformation/test_completion_extras_litellm_responses_transformation_transformation.py:7071
↓ 8 callersFunction_browser_session_cookie
( monkeypatch: pytest.MonkeyPatch, user_id: str, *, signing_key: str = _NATIVE_CLIENT_MASTER_K
tests/unit/proxy/mcp_server/test_discoverable_endpoints.py:11992
↓ 8 callersFunction_budget_id_of
(prisma: _FakePrisma, user_id: str, team_id: str = TEAM_ID)
tests/unit/proxy/management_endpoints/management_v1/test_teams.py:281
↓ 8 callersFunction_build_anthropic_tool_name_maps
Build (forward, reverse) tool-name maps for a single request. forward[original] = sanitized -- only present when name was rewritten reverse
litellm/llms/anthropic/chat/transformation.py:460
↓ 8 callersFunction_build_cluster_instance
()
tests/unit/caching/test_redis_cluster_node_isolation.py:88
↓ 8 callersFunction_build_db_connection_url_params
Build the Prisma DATABASE_URL query params controlling connection pool behavior. `connect_timeout` / `socket_timeout` map to the Prisma URL param
litellm/proxy/proxy_cli.py:71
↓ 8 callersMethod_build_decode_kwargs
Build the audience/issuer/options kwargs for ``jwt.decode``. Setting ``JWT_AUDIENCE`` (and optionally ``JWT_ISSUER``) turns on the co
litellm/proxy/auth/handle_jwt.py:1096
↓ 8 callersFunction_builtin_cost_map
()
litellm/utils.py:3886
↓ 8 callersFunction_cache
()
tests/unit/proxy/openai_files_endpoint/test_file_usage_caps.py:25
↓ 8 callersFunction_cache_async_client
(router: Router)
tests/unit/router_utils/test_kubernetes_pod_discovery.py:363
↓ 8 callersFunction_call
( proxy: Gateway, model: str, marker: str, metadata: JsonValue | None = None, key: str | N
tests/integration/observability/test_arize_otel_v2_openinference_sad_edge.py:31
↓ 8 callersFunction_call
(mode: Mode, **kwargs: object)
tests/integration/sdk/test_ollama_embedding_wire.py:62
↓ 8 callersFunction_call
(table: MagicMock, query: str, path: str = "/model-insights")
tests/unit/proxy/management_endpoints/test_model_insights_endpoints.py:80
↓ 8 callersFunction_call_summary_model
Invoke the configured summary model. Prefers ``llm_router.acompletion`` so the model alias resolves against the proxy's ``model_list``; falls
litellm/llms/anthropic/messages/context_management/editors/compact.py:507
↓ 8 callersFunction_call_transform_response
( handler: LiteLLMResponsesTransformationHandler, raw_response: "ResponsesAPIResponse", )
tests/unit/completion_extras/litellm_responses_transformation/test_completion_extras_litellm_responses_transformation_transformation.py:5094
↓ 8 callersFunction_callers_files_on_every_deployment
()
tests/unit/proxy/openai_files_endpoint/test_files_endpoint.py:6287
↓ 8 callersFunction_capture_login
(monkeypatch, on_login=lambda: None)
tests/unit/proxy/client/cli/test_up_commands.py:283
↓ 8 callersFunction_capture_scope_middleware
Middleware wired to a downstream that records the scope it received.
tests/unit/proxy/middleware/test_per_request_root_path_middleware.py:63
↓ 8 callersFunction_capture_token_increments
Answers the TTL-preserving token increment script and records main's ``(key, increment_value, ttl)``.
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:118
↓ 8 callersFunction_capturing_fetch
()
tests/unit/proxy/spend_tracking/test_background_interaction_settlement.py:190
↓ 8 callersFunction_card_stream
()
tests/unit/llms/bedrock/passthrough/guardrail_translation/test_handler.py:2596
↓ 8 callersFunction_carried
(error: Exception)
tests/integration/sdk/test_stream_open_error_provider_headers_sdk.py:214
↓ 8 callersFunction_carriers
(body: Mapping[str, JsonValue])
tests/integration/messages_endpoint/test_guardrail_long_tool_names_wire.py:109
↓ 8 callersFunction_carries
(span: Span, marker: str)
tests/integration/observability/test_otel_team_capture_message_content.py:285
↓ 8 callersFunction_carrying
(received: tuple[Request, ...], marker: str)
tests/integration/providers/test_device_code_login_guard_wire.py:383
↓ 8 callersFunction_chat
(gateway: Gateway, model: str, marker: str, *, key: str | None = None, **params: JsonValue)
tests/integration/providers/test_bedrock_runtime_chat_completions_sad_wire.py:63
↓ 8 callersFunction_chat
(rig: Rig, content: object, **extra: object)
tests/integration/observability/test_akto_logging_only.py:294
↓ 8 callersFunction_chat
(gateway: Gateway, body: Mapping[str, JsonValue], *, key: str | None = None)
tests/integration/caching/test_response_cache_empty_output.py:98
↓ 8 callersMethod_chat
(messages, params, model="openai/gpt-5.6")
tests/unit/integrations/test_anthropic_cache_control_hook.py:3577
↓ 8 callersFunction_chat_messages
(tool_contents: Sequence[JsonValue])
tests/integration/providers/test_gemini_tool_result_part_reference_wire.py:203
↓ 8 callersFunction_chat_status
(proxy: ProxyClient, key: str, body: BaseModel)
tests/e2e/llm_translation/test_chat_completions_contract_e2e.py:56
↓ 8 callersFunction_chatgpt_handler_kwargs
(caller_params, client)
tests/unit/llms/custom_httpx/test_llm_http_handler.py:567
↓ 8 callersFunction_chatgpt_sse_response
()
tests/unit/llms/custom_httpx/test_llm_http_handler.py:544
↓ 8 callersFunction_check_allowed_routes_caller_permission
Require PROXY_ADMIN when `allowed_routes` is present in the request body, unless the caller went through the `key_type` preset flow. Raw
litellm/proxy/management_endpoints/key_management_endpoints.py:1081
↓ 8 callersFunction_check_byok_credential
If the MCP server is BYOK-enabled, verify that the requesting user has a stored credential. When no credential is found, raise an HTTP 401 w
litellm/proxy/mcp_server/operations.py:1926
↓ 8 callersMethod_check_max_streaming_duration
Raise litellm.Timeout if the stream has exceeded LITELLM_MAX_STREAMING_DURATION_SECONDS.
litellm/litellm_core_utils/streaming_handler.py:530
↓ 8 callersMethod_check_rate_limits
Check rate limits using THREE-PHASE approach to prevent partial increments. Phase 1: Read-only check of ALL limits (no increments)
litellm/proxy/hooks/dynamic_rate_limiter_v3.py:372
↓ 8 callersFunction_classifier_rows
(rows: Sequence[Mapping[str, JsonValue]])
tests/integration/routing/test_complexity_router_databricks_classifier.py:251
↓ 8 callersFunction_clean
(day: str)
tests/integration/spend/test_daily_activity_aggregated_breakdowns.py:35
↓ 8 callersFunction_clear_prometheus_registry
()
tests/unit/integrations/test_prometheus_rate_limit_labels.py:490
↓ 8 callersFunction_cli_callback
( proxy: Gateway, graph: Graph, browser: httpx.Client, session: CliSession, hint: str )
tests/integration/authorization/test_sso_blank_identity.py:483
↓ 8 callersFunction_client
(role: LitellmUserRoles)
tests/unit/proxy/anthropic_endpoints/test_skills_endpoints.py:45
↓ 8 callersFunction_client
(captured: dict[str, object])
tests/unit/llms/tencent/image_edit/test_tencent_image_edit_transformation.py:22
↓ 8 callersFunction_client_job
(job_id: str, status: str, training_file: str = "file-abc123")
tests/integration/compatibility/test_fine_tuning_jobs.py:60
↓ 8 callersFunction_client_token
(account: str)
tests/unit/llms/chatgpt/chat/test_chatgpt_chat_transformation.py:283
↓ 8 callersFunction_cloudflare_config
()
tests/unit/llms/cloudflare/test_cloudflare_extraction_transformation.py:59
↓ 8 callersFunction_codex_request
A Responses API request shaped like what the Codex CLI sends when an MCP server is configured
tests/unit/llms/openai/responses/test_openai_responses_guardrail_handler.py:4398
↓ 8 callersFunction_cohere_embed_relay_result
( body: Mapping[str, object], native_path: str = "providers/cohere/v2/embed" )
tests/unit/llms/azure_ai/passthrough/test_azure_ai_passthrough_transformation.py:538
↓ 8 callersFunction_collected_samples
(metric_name: str)
tests/unit/integrations/test_prometheus_rate_limit_labels.py:500
↓ 8 callersFunction_complete
(router: Router, model: str, *, stream: bool, nonce: str | None = None, **request: object)
tests/integration/sdk/test_router_include_fallback_errors_wire.py:210
↓ 8 callersMethod_completion_response_text_parts
Collect non-empty text segments from chat, text completions, or responses API. Includes assistant message content *and* model-generated tool-
litellm/proxy/guardrails/guardrail_hooks/microsoft_purview/purview_dlp.py:245
↓ 8 callersFunction_concatenated_json
(*payloads: dict[str, object])
tests/unit/litellm_core_utils/prompt_templates/test_litellm_core_utils_prompt_templates_factory.py:5246
↓ 8 callersFunction_config
(provider: str = "github")
tests/integration/database/test_roi_observed.py:53
↓ 8 callersFunction_config
( tmp_path: Path, spend_logs_metadata_fields: Mapping[str, JsonValue] | None, *, model_list: t
tests/integration/spend/test_spend_logs_metadata_fields.py:118
↓ 8 callersFunction_config
(location: str | None = "us")
tests/unit/llms/vertex_ai/audio_transcription/test_vertex_ai_realtime_transformation.py:71
↓ 8 callersMethod_config
(self, tmp_path, general_settings)
tests/unit/proxy/test_proxy_cli.py:3859
↓ 8 callersFunction_configured_core
(make_core: MakeCore)
tests/unit/test_router/core/test_registry_characterization_model_info.py:167
↓ 8 callersFunction_configured_passthrough
(gateway: Gateway, scenario: Scenario, marker: str, target: str, *, auth: bool)
tests/integration/spend/test_passthrough_request_tags.py:106
↓ 8 callersFunction_configured_team_metadata_validator
(validator)
tests/unit/proxy/management_endpoints/test_team_endpoints.py:14159
↓ 8 callersFunction_consent
(*groups: TelemetryGroup)
tests/unit/telemetry/test_consent.py:87
↓ 8 callersFunction_content_filter_guardrail
(action: str, guardrail_cls=None, **guardrail_kwargs)
tests/unit/proxy/test_proxy_logging_hook_detection.py:208
↓ 8 callersFunction_convert_openai_response_to_mcp_result
Convert a litellm completion response to MCP CreateMessageResult. Args: response: The litellm ModelResponse. model_name: The
litellm/proxy/mcp_server/sampling_handler.py:661
↓ 8 callersFunction_convert_to_bedrock_tool_call_result
OpenAI message with a tool result looks like: { "tool_call_id": "tool_1", "role": "tool", "name": "get_current_weathe
litellm/litellm_core_utils/prompt_templates/factory.py:4516
↓ 8 callersFunction_cool_down
(router: Router, *deployment_ids: str)
tests/unit/test_router/test_router.py:6518
↓ 8 callersFunction_cost_the_logger_records_for
( url_route: str, *, stream: bool, reply: dict[str, object], stream_events: list[dict[str,
tests/unit/proxy/pass_through_endpoints/llm_provider_handlers/test_vertex_passthrough_logging_handler.py:140
↓ 8 callersFunction_count_body
(wire: Wire)
tests/integration/providers/test_openai_count_tokens_tools_wire.py:181
↓ 8 callersMethod_count_pre_call_check_tokens
``_count_pre_call_check_tokens``, which context compaction counts with.
litellm/router_rust/surface/members.py:112
↓ 8 callersFunction_count_request
(logger: PrometheusLogger, user_agent: str)
tests/unit/integrations/test_prometheus_series_cardinality.py:79
↓ 8 callersFunction_counted_body
( counter_type: type[AnthropicTokenCounter] | type[AzureAIAnthropicTokenCounter], count_endpoint: str,
tests/unit/llms/anthropic/test_anthropic_count_tokens_transformation.py:553
↓ 8 callersMethod_create
(self, request: FallbackCreateRequest, stored: _StoredRouterSettings)
tests/unit/proxy/test_fallback_management_endpoints.py:347
↓ 8 callersFunction_create_guardrail
(candidate: Gateway, identity: str, params: Mapping[str, JsonValue])
tests/integration/observability/_logging_only_scope_support.py:863
↓ 8 callersMethod_create_mock_httpx_response
Create a mock httpx response
tests/unit/proxy/pass_through_endpoints/llm_provider_handlers/test_openai_passthrough_logging_handler.py:69
↓ 8 callersFunction_create_team
(client: ManagementClient, resources: ResourceManager, alias: str, models: list[str])
tests/e2e/management/test_management_e2e.py:74
↓ 8 callersFunction_create_user
(client: ManagementClient, resources: ResourceManager, body: UserNewBody)
tests/e2e/management/test_management_e2e.py:80
↓ 8 callersFunction_created_key
(scenario: Scenario, created: dict[str, JsonValue])
tests/integration/management/test_key_default_allowed_routes.py:97
↓ 8 callersMethod_creator_turned_off_message_logging
Whether the creating key's or team's logging settings turn off message logging, falling back to the proxy-wide setting. A batch whose saved cr
enterprise/litellm_enterprise/proxy/common_utils/check_batch_cost.py:1097
↓ 8 callersFunction_current_json
(providers: Mapping[str, JsonValue])
tests/code_coverage_tests/test_generate_provider_endpoints_support.py:24
↓ 8 callersFunction_current_minute
()
tests/unit/test_router/test_io_token_rate_limits.py:1316
↓ 8 callersFunction_dashboard_ui_session_token
(user_id: str)
tests/integration/mcp/test_mcp_toolsets.py:39
← previousnext →8,501–8,600 of 202,479, ranked by callers