MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 15 callersFunction_response
( transcript: str | None, *, is_final: bool = False, billed: float = 0.0, event: str = "SP
tests/unit/llms/vertex_ai/audio_transcription/test_vertex_ai_realtime_backend.py:39
↓ 15 callersFunction_response_object
(response: httpx.Response)
tests/integration/mcp/test_mcp_oauth_flows.py:57
↓ 15 callersFunction_router
(redis_cache: FakeRedisCache, routing_strategy: str = "usage-based-routing-v2")
tests/unit/caching/test_request_redis_batch_pre_call.py:320
↓ 15 callersFunction_router_with_two_deployments
(blocked_flags)
tests/unit/test_router/test_router.py:6325
↓ 15 callersFunction_run_dispatch
(wrapper: CustomStreamWrapper, chunk)
tests/unit/litellm_core_utils/test_streaming_handler.py:4400
↓ 15 callersMethod_run_shadow_eval
Budget gates once per sampled request, then every router arm in turn: shadow call -> blind judge -> one attempt row stamped with the arm. The
litellm/integrations/shadow_eval_logger.py:1095
↓ 15 callersFunction_run_streaming_step
(translation, streaming_chunks=None)
tests/unit/proxy/policy_engine/test_pipeline_executor.py:1249
↓ 15 callersFunction_run_team_trace_id_check
( *, route: str, method: str = "POST", request_body: dict[str, object] | None = None, head
tests/unit/proxy/auth/test_auth_checks.py:56
↓ 15 callersFunction_safe_output
(output: str)
tests/integration/database/test_v1_migration_error_logging.py:210
↓ 15 callersFunction_sample_agent_response
( agent_id: str = "agent-123", agent_name: str = "Test Agent" )
tests/unit/proxy/agent_endpoints/test_endpoints.py:81
↓ 15 callersFunction_scripted
(scenario: Scenario)
tests/integration/providers/test_openai_realtime_transcription_url.py:129
↓ 15 callersFunction_send_through_proxy
( path: str, headers: Mapping[str, str], method: str = "POST" )
tests/unit/proxy/pass_through_endpoints/test_pass_through_endpoints.py:11611
↓ 15 callersFunction_sent
Subset of the call kwargs limited to `keys`, for exact comparison.
tests/unit/batches/test_main.py:488
↓ 15 callersFunction_session_update
(audio_input: dict[str, object])
tests/unit/llms/soniox/realtime/test_soniox_realtime_transformation.py:41
↓ 15 callersMethod_set_langfuse_specific_attributes
Sets Langfuse specific metadata attributes onto the OTEL span. All keys supported by the vanilla Langfuse integration are mapped to
litellm/integrations/langfuse/langfuse_otel.py:289
↓ 15 callersFunction_shared_cache
A DualCache whose replay guard is backed by a shared, atomic store. An InMemoryCache instance stands in for Redis; passing the same instance to
tests/unit/proxy/management_endpoints/test_saml_sso.py:47
↓ 15 callersFunction_shield_response
(attack_detected)
tests/unit/proxy/guardrails/guardrail_hooks/azure/test_azure_prompt_shield.py:274
↓ 15 callersFunction_sign_id_token
(claims: Mapping[str, object])
tests/unit/proxy/mcp_server/test_oauth_identity_binding.py:47
↓ 15 callersFunction_signed_jwt
( private_key: rsa.RSAPrivateKey, client_id: str, issuer: str | None = None, *, subject: s
tests/integration/authorization/test_jwt_key_mapping_lifecycle.py:218
↓ 15 callersFunction_spend_log_row
(digest: str, key_alias: str | None, team_id: str | None, user_id: str | None)
tests/unit/proxy/spend_tracking/test_key_metadata_recovery.py:54
↓ 15 callersFunction_stream
(rig: Rig, guardrail: str, marker: str)
tests/integration/observability/test_bedrock_converse_stream_post_call_guardrail.py:442
↓ 15 callersMethod_success_handler_helper_fn
( self, result: object = None, start_time: datetime.datetime | None = None, en
litellm/litellm_core_utils/litellm_logging.py:3309
↓ 15 callersFunction_summary
(marker: str)
tests/integration/providers/test_responses_compaction_items_wire.py:83
↓ 15 callersFunction_team_member
( team_id: str = "team1", models: list[str] | None = None, **aliases: dict[str, str] | None )
tests/unit/proxy/test_model_list_aliases.py:74
↓ 15 callersFunction_team_rows
(team_id: str)
tests/integration/management/test_team_delete_large_membership.py:85
↓ 15 callersMethod_template
(self, body: dict)
tests/unit/llms/sap/chat/test_sap_transformation.py:662
↓ 15 callersMethod_to_model_list
Convert a list of database records to domain models.
litellm/repositories/base_repository.py:75
↓ 15 callersFunction_tools
(body: Mapping[str, JsonValue], invoke: bool)
tests/integration/providers/test_bedrock_native_structured_output_wire.py:205
↓ 15 callersFunction_transcribe
( ws_base: str, query: str, key: str | None, *, path: str = "/v1/realtime", update: Js
tests/integration/providers/test_realtime_transcription_guardrail_session_update.py:625
↓ 15 callersFunction_transport
( pull_detail_status: int = 200, unexpected_details: bool = False, profile_email: str = "alice@exa
tests/unit/proxy/roi_calculator/test_sync.py:179
↓ 15 callersFunction_unarmed_logging_obj
Real Logging objects only carry _on_deferred_stream_complete when the proxy arms deferred dispatch; a bare MagicMock's auto-attribute is truthy
tests/unit/proxy/pass_through_endpoints/test_streaming_handler_interrupt.py:44
↓ 15 callersMethod_update_headers_with_anthropic_beta
Auto-inject anthropic-beta headers based on features used. Handles: - context_management: adds 'context-management-2025-06-2
litellm/llms/anthropic/messages/native/transformation.py:852
↓ 15 callersFunction_upstream
(pause: float = 0)
tests/integration/messages_endpoint/chat_bridge/test_chat_bridge_multi_tool_call_blocks_wire.py:383
↓ 15 callersFunction_validation_cache
()
tests/unit/proxy/auth/test_auth_checks_object_access_and_lookup.py:6523
↓ 15 callersFunction_verdict
( master_key: str | None, general_settings: Mapping[str, object] | None = None, *, environment
tests/unit/proxy/auth/test_master_key_boot_check.py:39
↓ 15 callersFunction_verdict
(efficient: float = 0.90, capable: float = 0.92)
tests/unit/router_strategy/test_llm_v2.py:50
↓ 15 callersFunction_voice_scenario
(*transcripts: JsonValue)
tests/integration/providers/test_realtime_transcription_guardrail_session_update.py:341
↓ 15 callersFunction_wire_team_delete_tx
`/team/delete` records, deletes and sweeps the team rows under every team's advisory lock in one transaction, so a mocked client has to hand its o
tests/unit/proxy/management_endpoints/test_team_endpoints.py:228
↓ 15 callersMethod_write
(self, change: Callable[[set[object]], object])
litellm/router_rust/surface/indexes.py:209
↓ 15 callersFunctiona
()
litellm/proxy/_experimental/out/_next/static/chunks/2k-fkubkt19cc.js:8
↓ 15 callersMethodaccess_token
Sign `identity` in through the direct-access grant and hand back the access token Keycloak signed, exactly as it came off the wire.
tests/e2e/idp.py:275
↓ 15 callersMethodacompletion
(self, **_: object)
tests/unit/proxy/guardrails/guardrail_hooks/litellm_content_filter/test_content_filter.py:2012
↓ 15 callersMethodadd_exact_path_route
Add exact path route for pass-through endpoint
litellm/proxy/pass_through_endpoints/pass_through_endpoints.py:3853
↓ 15 callersMethodadd_litellm_success_callback
Add a success callback to `litellm.success_callback`. Auto-routes async callbacks to litellm._async_success_callback. Special
litellm/litellm_core_utils/logging_callback_manager.py:78
↓ 15 callersFunctionadd_provider_specific_headers_to_request
( data: dict, headers: Mapping[str, str], )
litellm/proxy/litellm_pre_call_utils.py:4488
↓ 15 callersMethodadd_server
(self, mcp_server: LiteLLM_MCPServerTable)
litellm/proxy/mcp_server/mcp_server_manager.py:3487
↓ 15 callersMethodadd_user_api_key_auth_to_request_metadata
Adds the `UserAPIKeyAuth` object to the request metadata.
litellm/proxy/litellm_pre_call_utils.py:2117
↓ 15 callersFunctionadmit_managed_actor
(auth: UserAPIKeyAuth, store: AgentIdentityStore | None)
litellm/proxy/agent_endpoints/auth/managed_authorization.py:159
↓ 15 callersMethodafile_delete
( self, model: str, file_id: str, litellm_metadata: Mapping[str, object],
tests/unit/enterprise/proxy/test_managed_files_hook.py:2289
↓ 15 callersFunctionagents_table
(prisma_client: PrismaClient)
litellm/proxy/agent_endpoints/agent_registry.py:133
↓ 15 callersMethodanswers
(self)
tests/integration/providers/_assistants_route_errors_support.py:418
↓ 15 callersMethodapply_client_tag_policy_pre_auth
Merge ``x-litellm-tags`` header tags into ``request_data`` BEFORE auth budget gates run, so ``_tag_max_budget_check`` (which only
litellm/proxy/litellm_pre_call_utils.py:2525
↓ 15 callersMethodapply_guardrail
Apply Prompt Security guardrail to the given inputs. This method is called by LiteLLM's guardrail framework for ALL endpoints:
litellm/proxy/guardrails/guardrail_hooks/prompt_security/prompt_security.py:177
↓ 15 callersFunctionapply_policies
Apply guardrails to inputs from policy names and/or a direct list of guardrail names. Runs all guardrails in order; if one fails, the error
litellm/proxy/management_endpoints/policy_endpoints/endpoints.py:122
↓ 15 callersMethodapply_runtime_values
(self, values: Mapping[str, JsonValue])
litellm/proxy/config_resolvers/settings_store.py:103
↓ 15 callersMethodapply_to_anthropic_messages_request
Apply cache control injection for the Anthropic-native v1/messages endpoint. ``external_breakpoints`` is the client's breakpoint count outsid
litellm/integrations/anthropic_cache_control_hook.py:1387
↓ 15 callersMethodarm
(self)
tests/integration/_support/database_relay.py:30
↓ 15 callersFunctionassert_enforced_within
(elapsed: float, *, configured: float, slack: float, not_before: float = 0.0)
tests/integration/_support/timing.py:33
↓ 15 callersMethodasync_get_cache
- Check if the cache key is in-memory - Else: - add missing cache key from REDIS - update in-memory cache
litellm/proxy/hooks/batch_redis_get.py:86
↓ 15 callersMethodasync_get_token
(self, config: DatabricksAppOAuthConfig)
litellm/proxy/agent_endpoints/databricks_oauth.py:231
↓ 15 callersFunctionasync_inline_remote_media
( messages: list[AllMessageValues], # mutable-ok: every transform_request takes list[AllMessageValues]
litellm/litellm_core_utils/prompt_templates/image_handling.py:553
↓ 15 callersMethodasync_read_secret
Async function to read a secret from AWS Secrets Manager Returns: str: Secret value Raises: ValueErr
litellm/secret_managers/aws_secret_manager_v2.py:137
↓ 15 callersMethodasync_write_secret
Async function to write a secret to AWS Secrets Manager Args: secret_name: Name of the secret secret_value:
litellm/secret_managers/aws_secret_manager_v2.py:267
↓ 15 callersFunctionat
(clock: &Clock, seconds: u64)
litellm-rust/crates/cache-memory/tests/cache.rs:36
↓ 15 callersFunctionauthorize
( request: Request, redirect_uri: str, client_id: str | None = None, state: str = "", mcp_
litellm/proxy/mcp_server/discoverable_endpoints.py:2368
↓ 15 callersFunctionavailable
(tool_uses: Mapping[str, JsonValue])
tests/integration/_support/safeguards_rig.py:390
↓ 15 callersFunctionazure_storage_config
( path: Path, settings: Mapping[str, JsonValue] | None = None, *, callback_setting: str = "callbacks" )
tests/integration/observability/_azure_storage_support.py:177
↓ 15 callersFunctionbase_url
(candidate: Gateway)
tests/integration/authorization/_hidden_alias_budget.py:94
↓ 15 callersFunctionbudget_row
A fake litellm_budgettable row whose model_dump returns the given fields.
tests/unit/proxy/common_utils/test_upsert_budget_membership.py:46
↓ 15 callersMethodbuild_response_from_payload
( self, payload: dict[str, object], # mutable-ok: mirrors override contract model_res
litellm/llms/soniox/audio_transcription/transformation.py:426
↓ 15 callersFunctioncache_key
(input: &CacheKeyInput)
litellm-rust/crates/cache-response/src/caching.rs:76
↓ 15 callersFunctioncallback
OAuth 2.0 authorization response handler for MCP loopback clients. Accepts either: - A successful authorization response (``code`` + ``state
litellm/proxy/mcp_server/discoverable_endpoints.py:2679
↓ 15 callersFunctioncallback
(*args: object)
tests/unit/test_lint_base_counts.py:169
↓ 15 callersMethodcalls
(self, marker: str)
tests/integration/routing/test_non_chat_model_endpoint_gate.py:115
↓ 15 callersFunctioncanary_rig
( root: Path, *, configure: Callable[[dict[str, object], str], None] | None = None, environmen
tests/integration/security/_sinks.py:208
↓ 15 callersFunctionchange_password
Change the calling user's own password. Only callable with the dashboard session issued by a username/password login; SSO sessions and v
litellm/proxy/management_endpoints/password_endpoints.py:76
↓ 15 callersMethodchat
( self, marker: str, *, headers: Mapping[str, str] | None = None, key:
tests/integration/observability/test_otel_conversation_id.py:225
↓ 15 callersFunctionchat_override
`chat_turns_override` for the single user turn most reliability tests send.
tests/e2e/router/reliability_support.py:327
↓ 15 callersMethodchat_status
(self, key: str, model: str, content: str)
tests/e2e/management/management_client.py:711
↓ 15 callersFunctioncheck_disable_global_guardrails_caller_permission
Only proxy admins may opt a key or team out of default-on guardrails, whether the flag is top-level or under `metadata`. Re-sending a flag th
litellm/proxy/management_endpoints/common_utils.py:245
↓ 15 callersMethodclassify
Classify a prompt by complexity, discarding which rule decided the tier. Kept for callers that only need the tier and score; `_score_and_clas
litellm/router_strategy/complexity_router/complexity_router.py:1829
↓ 15 callersMethodcleanup
(self)
tests/integration/_support/generation.py:34
↓ 15 callersMethodclear
(&self)
litellm-rust/crates/cache-response/src/buffer.rs:41
↓ 15 callersMethodcli_on_every_node
(&self, args: &[&str])
litellm-rust/crates/redis-testing/src/cluster.rs:86
↓ 15 callersMethodclose
(self)
tests/integration/streaming/test_assistants_evals_stream.py:155
↓ 15 callersFunctioncompacts
(requests: tuple[Upstreamed, ...])
tests/integration/authorization/_responses_compact.py:202
↓ 15 callersFunctionconfigure_codex_settings
( base_url: str, api_key: str, model: str, settings_path: Path, *, commit: Callable[[s
litellm/proxy/client/cli/commands/codex_settings.py:188
↓ 15 callersMethodconnect
(self)
litellm/proxy/utils.py:7023
↓ 15 callersFunctionconnect_observed
(deployment: &RedisDeployment)
litellm-rust/crates/redis-store/tests/support/mod.rs:64
↓ 15 callersFunctioncontext_window
(model: str = "gpt-4o")
tests/unit/test_router/core/scripted_host.py:129
↓ 15 callersMethodcount
(self, *, where: Mapping[str, object])
litellm/repositories/prisma_protocols.py:126
↓ 15 callersFunctioncredential_less_proxy
An operator with no Langfuse account and no generic OTLP collector.
tests/unit/integrations/otel/test_otel_v2_destinations.py:2948
↓ 15 callersFunctioncurrent_test_key
The pytest node id of the running test, from the PYTEST_CURRENT_TEST env var pytest maintains (``<nodeid> (setup|call|teardown)``); ``session`` fo
tests/e2e/fixture_mode.py:55
↓ 15 callersMethodcursors
(self)
tests/integration/observability/test_otel_team_capture_message_content.py:188
↓ 15 callersFunctiondelete_nested_value
Delete a field from nested data using JSONPath notation. Custom implementation - no external dependencies. Supports: - "field" - to
litellm/litellm_core_utils/dot_notation_indexing.py:183
↓ 15 callersFunctiondelta_text
(events: tuple[SseEvent, ...])
tests/integration/_support/anthropic_sse.py:165
↓ 15 callersFunctiondetect_stalled_task
Whether the newest tool call is still part of a stuck pattern: it repeats, or it errored, at least repeat_threshold times across the last `window`
litellm/router_strategy/complexity_router/stall_detector.py:95
↓ 15 callersMethoddirty
()
litellm/proxy/_experimental/out/_next/static/chunks/2j7hjtph8n74b.js:1
↓ 15 callersFunctiondownload
( candidate: Gateway, key: str, file_id: str, *, route: str = "/v1/files/{}/content",
tests/integration/management/_batch_file_caps.py:221
↓ 15 callersFunctiondownloads_seen
(requests: tuple[Request, ...], file_id: str)
tests/integration/management/_batch_file_caps.py:156
← previousnext →4,001–4,100 of 202,479, ranked by callers