MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 5 callersFunctionadd_object_type
(schema)
litellm/llms/vertex_ai/common_utils.py:1079
↓ 5 callersFunctionadd_retry_headers_to_response
Add retry headers to the request
litellm/litellm_core_utils/add_retry_fallback_headers.py:306
↓ 5 callersFunctionadd_upstream_authorization
(gateway: Gateway, peer: McpPeer, key: str, identity: str)
tests/integration/mcp/test_mcp_user_env_vars.py:97
↓ 5 callersFunctionadmin_can_invite_user
Check if the admin can create an invitation for the target user. - Proxy admins: can invite any user - Org admins: can only invite users
litellm/proxy/management_endpoints/common_utils.py:456
↓ 5 callersFunctionadministersTeam
(team: TeamMembers | null, userId: string | null)
ui/litellm-dashboard/src/components/organisms/createKeyOwnerAccess.ts:21
↓ 5 callersFunctionadmit
(call_id: str)
tests/unit/proxy/test_common_request_processing.py:13121
↓ 5 callersMethodadmit
Return the number of timeouts suppressed since the last admitted line, or None to suppress this one.
litellm/caching/redis_cache.py:386
↓ 5 callersMethodadmit
(&self, now: Instant)
litellm-rust/crates/circuit-breaker/src/breaker.rs:149
↓ 5 callersMethodadmit_series
(self, metric_name: str, label_values: tuple[str, ...], max_series: int)
litellm/integrations/prometheus_helpers/shared_prometheus_series_admissions.py:85
↓ 5 callersMethodadvance
(self, seconds: float)
tests/unit/router_utils/test_prompt_caching_cache.py:24
↓ 5 callersMethodadvance
(self, seconds: float)
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:95
↓ 5 callersMethodadvance
(self, seconds: float)
tests/unit/proxy/hooks/test_model_max_budget_limiter.py:218
↓ 5 callersMethodaembed
(self, model: str, query: str, configuration: Mapping[str, object])
litellm/llms/base_llm/vector_store/transformation.py:41
↓ 5 callersMethodaembed
(self, model: str, query: str, configuration: Mapping[str, object])
litellm/llms/base_llm/vector_store/transformation.py:115
↓ 5 callersMethodafile_content
( self, file_content_request: FileContentRequest, openai_client: AsyncOpenAI, )
litellm/llms/openai/openai.py:1867
↓ 5 callersMethodafter_provider_response
(&self, raw: RawResponse)
litellm-rust/crates/host-http/tests/serve.rs:168
↓ 5 callersMethodagent
(self, agent_id: str)
litellm/proxy/agent_endpoints/identity_store.py:70
↓ 5 callersFunctionagent_caller_auth
A minimal auth context standing for the invoking user and team, so the shared key/team/user resolvers can be reused unchanged to compute what the
litellm/proxy/agent_endpoints/auth/agent_caller.py:39
↓ 5 callersMethodagrep
( self, pattern: str, path: str | None = None, glob: str | None = None,
litellm/llms/deepagents/harness/sandbox_backend.py:340
↓ 5 callersMethodaimage_edit
( self, model: str, image: Any, prompt: str, model_response: ImageResp
tests/unit/llms/test_custom_llm.py:318
↓ 5 callersMethodalerting_handler
(self, message: str, level: str, alert_type: object)
tests/integration/database/test_spendlogs_ledger_parity.py:141
↓ 5 callersMethodalias
(self, name: str)
litellm/integrations/ternary/export_engine.py:33
↓ 5 callersMethodalias
(self, cell: Cell)
tests/e2e/llm_translation/conversational_matrix.py:215
↓ 5 callersFunctionall_deployments
`_core_get_all_deployments(model_name, team_id)`: the team's deployments under that public name when it has any, else the deployments named `model_nam
litellm-rust/crates/router-core/src/resolve.rs:59
↓ 5 callersFunctionall_fields_present
(standard_logging_metadata: StandardLoggingMetadata)
tests/unit/litellm_core_utils/test_litellm_logging.py:12554
↓ 5 callersFunctionallowlisted
(value: object, allowed: frozenset[str])
litellm/litellm_core_utils/bug_report.py:74
↓ 5 callersFunctionan
(t,e,r)
litellm/proxy/_experimental/out/_next/static/chunks/0cz1d0mv5g_q7.js:1
↓ 5 callersFunctionanswer
(marker: str | None)
tests/integration/_support/responses_vendor.py:35
↓ 5 callersMethodanthropic_sse_wrapper
Convert AnthropicStreamWrapper dict chunks to Server-Sent Events format. Similar to the Bedrock bedrock_sse_wrapper implementation.
litellm/llms/anthropic/messages/adapters/streaming_iterator.py:1370
↓ 5 callersFunctionanthropic_tool_search_chat_tool
()
litellm/responses/litellm_completion_transformation/tool_search.py:120
↓ 5 callersFunctionapi_base
(request: NativeCall)
litellm/rust_bridge/detectors.py:99
↓ 5 callersFunctionapp
(span: Mapping[str, object])
tests/integration/observability/test_langfuse_otel_metadata_destinations.py:262
↓ 5 callersFunctionapp
Return the proxy_server FastAPI app with lifespan effectively disabled. TestClient used WITHOUT the ``with`` context manager skips the lifespan,
tests/unit/proxy/proxy_server/conftest.py:76
↓ 5 callersFunctionapplyKeyUpdate
(t *testing.T, client *Client, stateAttrs map[string]string, config map[string]interface{})
terraform/provider/litellm/resource_key_test.go:1111
↓ 5 callersFunctionapplyTierSetAction
( value: ComplexityRouterConfigValue, keywordTierRules: readonly KeywordTierRule[], action: TierSetActio
ui/litellm-dashboard/src/components/add_model/tier_set_actions.ts:159
↓ 5 callersFunctionapply_delta
Apply a learning update to a cell, enforcing the sample cap. SAMPLE_CAP is a HARD cap on (alpha + beta). When the cap would be exceeded,
litellm/router_strategy/adaptive_router/bandit.py:64
↓ 5 callersMethodapply_guardrail
( self, inputs: GenericGuardrailAPIInputs, request_data: dict[str, object], # mutable
litellm/proxy/guardrails/guardrail_hooks/decision_model/decision_model.py:308
↓ 5 callersMethodapply_guardrail
( self, inputs: GenericGuardrailAPIInputs, request_data: dict, input_type: Lit
litellm/proxy/guardrails/guardrail_hooks/xecguard/xecguard.py:163
↓ 5 callersMethodapply_guardrail
Unified guardrail method for the apply_guardrail framework. Called by the UI "Test Guardrail" endpoint, UnifiedLLMGuardrails orchest
litellm/proxy/guardrails/guardrail_hooks/panw_prisma_airs/panw_prisma_airs.py:1850
↓ 5 callersFunctionapply_off_peak_pricing
Swap in off-peak per-token rates when the current UTC time is inside one of the model's off_peak_pricing rules, the every-day hours_utc windows or
litellm/litellm_core_utils/llm_cost_calc/utils.py:672
↓ 5 callersMethodapply_stream_scope
(self, stream_scope: object)
litellm/integrations/custom_guardrail.py:1389
↓ 5 callersFunctionarchive_source
(name: str, digest: str = "a" * 64)
tests/integration/management/test_claude_code_plugin_installation_preference.py:41
↓ 5 callersMethodargument
( &self, py: Python<'py>, arguments: &Bound<'py, PyDict>, name: &str, )
litellm-rust/crates/python-bridge/src/routes/mod.rs:88
↓ 5 callersFunctionariaOf
(control: MountedFieldControlProps)
ui/litellm-dashboard/src/app/(dashboard)/mcp-servers/_components/mcpFieldRules.ts:11
↓ 5 callersMethodarun_thread
( self, thread_id: str, assistant_id: str, additional_instructions: str | None
litellm/llms/openai/openai.py:3173
↓ 5 callersFunctionasString
(value: unknown)
ui/litellm-dashboard/src/components/logs/detail/prettyMessagesUtils.ts:20
↓ 5 callersFunctionas_converted_stream
( params: Mapping[str, object], prefix: str )
litellm/types/integrations/custom_logger.py:53
↓ 5 callersMethodas_dict
(self)
litellm/proxy/management_endpoints/credential_migration.py:94
↓ 5 callersFunctionas_router
(core: RouterCore)
tests/unit/test_router/core/test_gaps_router_core_crate.py:31
↓ 5 callersFunctionasgi_scope_fields
(scope: ASGIScopeFields)
litellm/proxy/mcp_server/utils.py:40
↓ 5 callersFunctionassert_batch_object
(batch: BatchObject)
tests/e2e/batches/test_batches_e2e.py:238
↓ 5 callersFunctionassert_compacted
(compacted: CompactedResponse, model: str, text: str)
tests/integration/authorization/_responses_compact.py:308
↓ 5 callersFunctionassert_count_tokens_shape
Return None on success, or an error string describing the first violation. Acceptance criteria are intentionally minimal: 1. The call succ
tests/e2e/claude_code/http_probe.py:355
↓ 5 callersMethodassert_current
(self, server: MCPServer)
litellm/proxy/mcp_server/catalog.py:178
↓ 5 callersFunctionassert_drained
(cache: OwnedRedis, group: str)
tests/integration/routing/test_priority_scheduler_queue_cleanup.py:753
↓ 5 callersFunctionassert_failed_once
(wire: Wire, call_id: str, model: str, attempts: int = 1)
tests/integration/streaming/test_responses_bridge_stream_errors.py:56
↓ 5 callersFunctionassert_limit_message
(body: str, descriptor: str, group: str, kind: str, limit: int)
tests/integration/routing/test_per_model_rate_limit_fallbacks.py:281
↓ 5 callersFunctionassert_names_deployment
(alert: str, api_base: str, model_group: str, backend: str, caller: Caller)
tests/integration/observability/test_stream_error_alert_api_base.py:146
↓ 5 callersMethodassert_no_provider_call
(self)
tests/unit/proxy/fine_tuning_endpoints/test_endpoints.py:113
↓ 5 callersFunctionassert_original_proof
(database: Database, row_id: str, finished: bool)
tests/e2e/migrations/checks.py:75
↓ 5 callersFunctionassert_priced_row
(row: Mapping[str, JsonValue], model: str)
tests/integration/providers/_mantle_gpt_prompt_cache_support.py:633
↓ 5 callersMethodassert_provider_result
(&self)
litellm-rust/crates/inference-messages/tests/live/support.rs:31
↓ 5 callersFunctionassert_real_ttft
The enforced behavior: the gen-AI span for the attempt that served the stream records a TTFT that is a real measurement - present, numeric, po
tests/e2e/logging/test_otel_trace_e2e.py:140
↓ 5 callersFunctionassert_signed
(request: Request, access_key: str, region: str, session_token: str | None)
tests/integration/providers/_sagemaker_chat_container.py:163
↓ 5 callersFunctionassert_sigv4
(requests: tuple[ControlPlaneRequest, ...], *, key: str, secret: str, region: str)
tests/integration/_support/bedrock_discovery.py:254
↓ 5 callersFunctionassert_user_can_access_container
( container_id: str, user_api_key_dict: UserAPIKeyAuth, custom_llm_provider: str, )
litellm/proxy/container_endpoints/ownership.py:328
↓ 5 callersFunctionassert_user_can_access_vector_store
Raise 403 unless the caller can access the resolved vector store.
litellm/proxy/vector_store_endpoints/utils.py:326
↓ 5 callersFunctionassistant_end
(text: str, stop_reason: str, error: str | None = None)
tests/unit/llms/pi/harness/test_transformation.py:81
↓ 5 callersFunctionassistant_turn
(text: str)
tests/integration/providers/test_responses_codex_repeated_additional_tools_wire.py:39
↓ 5 callersMethodasync_add_cache
Async implementation of add_cache
litellm/caching/caching.py:822
↓ 5 callersMethodasync_batch_set_cache
( self, cache_list: list[tuple[str, object]], litellm_parent_otel_span: Span | None,
litellm/proxy/utils.py:530
↓ 5 callersMethodasync_callback_filter_deployments
``async_callback_filter_deployments``: each routing callback narrows the deployments in turn.
litellm/router_rust/surface/legacy_members.py:369
↓ 5 callersFunctionasync_convert_url_to_base64
(url: str, timeout: float | httpx.Timeout | None = None)
litellm/litellm_core_utils/prompt_templates/image_handling.py:108
↓ 5 callersMethodasync_delete_cache
(self, key: str)
litellm/proxy/mcp_server/outbound_credentials/dual_cache_token_backend.py:33
↓ 5 callersMethodasync_deployment_callback_on_failure
Count a failed call as one more request of the minute (``async_deployment_callback_on_failure``).
litellm/router_rust/surface/legacy_members.py:483
↓ 5 callersMethodasync_filter_deployments
( self, model: str, healthy_deployments: list[dict[str, object]], # mutable-ok: Custo
litellm/router_utils/pre_call_checks/continuation_prefill_check.py:109
↓ 5 callersMethodasync_filter_deployments
Filter out deployments that have exceeded their provider budget limit. Example: if deployment = openai/gpt-3.5-turbo
litellm/router_strategy/budget_limiter.py:202
↓ 5 callersMethodasync_flush_passthrough_collected_chunks
Log the response the collector built from the stream, and say whether there was one to log.
litellm/litellm_core_utils/litellm_logging.py:3599
↓ 5 callersMethodasync_get_available_deployments
Returns a deployment with the lowest cost
litellm/router_strategy/lowest_cost.py:208
↓ 5 callersMethodasync_get_chat_completion_prompt
( self, model, messages, non_default_params, promp
tests/unit/litellm_core_utils/test_litellm_logging.py:9934
↓ 5 callersMethodasync_get_token
Return a valid access token, fetching or refreshing as needed. Returns ``None`` when the server lacks client credentials config.
litellm/proxy/mcp_server/oauth2_token_cache.py:96
↓ 5 callersMethodasync_health_check
Performs a health check for Arize integration. Returns: dict: Health check result with status and message
litellm/integrations/arize/arize.py:258
↓ 5 callersMethodasync_log_audit_log_event
Batch audit logs and upload to S3 under audit_logs/ prefix.
litellm/integrations/s3_v2.py:525
↓ 5 callersMethodasync_log_background_interaction_completion
Log the terminal result of a background interaction as a fresh success event. The create request already ran success logging for its
litellm/litellm_core_utils/litellm_logging.py:3530
↓ 5 callersMethodasync_log_failure_event
(self, kwargs, response_obj, start_time, end_time)
tests/unit/test_router/test_router.py:2691
↓ 5 callersMethodasync_log_success_event
( self, kwargs: Mapping[str, object], response_obj: object, start_time: object
litellm/integrations/galileo.py:669
↓ 5 callersMethodasync_log_success_event
( self, kwargs: Mapping[str, object], response_obj: object, start_time: dateti
litellm/integrations/otel/logger.py:360
↓ 5 callersMethodasync_log_success_event
Original method now uses helper functions
litellm/router_strategy/budget_limiter.py:570
↓ 5 callersMethodasync_moderation_hook
Runs in parallel to LLM API call Runs on only Input This can NOT modify the input, only used to reject or accept a call befo
litellm/proxy/guardrails/guardrail_hooks/unified_guardrail/unified_guardrail.py:322
↓ 5 callersMethodasync_post_call_success_hook
( self, data: dict[str, object], user_api_key_dict: UserAPIKeyAuth, response: LLMResponseTypes )
tests/unit/integrations/test_custom_guardrail_post_call_once.py:146
↓ 5 callersMethodasync_pre_call_hook
( self, user_api_key_dict: UserAPIKeyAuth, cache: DualCache, data: dict[str, o
tests/unit/integrations/vector_store_integrations/test_vector_store_pre_call_hook.py:765
↓ 5 callersMethodasync_rotate_secret
( &self, current_name: &str, new_name: &str, value: &SecretValue, )
litellm-rust/crates/secrets-cyberark/src/secret_manager/write.rs:154
↓ 5 callersMethodasync_routing_strategy_pre_call_checks
``async_routing_strategy_pre_call_checks``: a check's rate limit cools the deployment down under the Python rules (``set_cooldown_deployments`
litellm/router_rust/surface/legacy_members.py:341
↓ 5 callersMethodasync_rpush_pipeline
Use Redis Pipelines for bulk RPUSH operations Args: rpush_list: List of RedisPipelineRpushOperation dicts containing:
litellm/caching/redis_cache.py:2194
↓ 5 callersMethodasync_scan_request_attachments
( self, data: dict[str, object], # mutable-ok: proxy request body dict, mutated by the loggin
litellm/proxy/guardrails/guardrail_hooks/bedrock_guardrails.py:2564
↓ 5 callersMethodasync_send_batch
(self)
litellm/integrations/datadog/datadog_cost_management.py:95
↓ 5 callersMethodasync_send_user_invitation_email
Send a user invitation email to the user
litellm/proxy/hooks/user_management_event_hooks.py:96
↓ 5 callersMethodasync_set_cache
(self, key: str, value: float, ttl: int | None = None, **kwargs: object)
tests/unit/proxy/hooks/test_model_max_budget_limiter.py:244
↓ 5 callersMethodasync_set_cache_pipeline
( &self, entries: Vec<(String, Self::Value)>, context: Self::Context, )
litellm-rust/crates/cache-s3/src/cache.rs:212
↓ 5 callersMethodasync_should_run_agentic_loop
( self, response: object, model: str, messages: list[dict], tools: lis
litellm/integrations/code_interpreter_interception/handler.py:443
← previousnext →15,401–15,500 of 202,479, ranked by callers