MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 9 callersMethodload_local_beta_headers_config
Load the local backup beta headers config bundled with the package.
litellm/anthropic_beta_headers_manager.py:65
↓ 9 callersMethodlog_pre_api_call
(self, model: str, messages: object, kwargs: Mapping[str, object])
litellm/router_strategy/least_busy.py:425
↓ 9 callersMethodlogging_non_streaming_response
( self, model: str, custom_llm_provider: str, httpx_response: Response,
litellm/llms/vllm/passthrough/transformation.py:75
↓ 9 callersMethodlogs
(self)
tests/e2e/migrations/containers.py:62
↓ 9 callersFunctionmake_logger_with_budget_metrics_disabled
()
tests/unit/integrations/test_prometheus_budget_metric_guard.py:47
↓ 9 callersMethodmake_model_armor_request
Make request to Model Armor API. Supports both text and file prompt sanitization. If file_bytes and file_type are provided, file prom
litellm/proxy/guardrails/guardrail_hooks/model_armor/model_armor.py:341
↓ 9 callersFunctionmake_request
( *, headers: Optional[Dict[str, str]] = None, client: Optional[Tuple[str, int]] = ("203.0.113.7",
tests/unit/proxy/auth/test_network.py:16
↓ 9 callersFunctionmanager
(server: &MockServer, always_read: bool, ttl: Duration)
litellm-rust/crates/secrets-google/tests/secret_manager.rs:12
↓ 9 callersMethodmap_openai_params
Map OpenAI parameters to Moonshot AI parameters Handles Moonshot AI specific limitations: - tool_choice doesn't support "req
litellm/llms/moonshot/chat/transformation.py:120
↓ 9 callersFunctionmarkers
(outcomes: Sequence[Outcome])
tests/integration/routing/test_payment_required_cooldown.py:238
↓ 9 callersFunctionmask_sensitive_structure
(data: object)
litellm/litellm_core_utils/sensitive_data_masker.py:173
↓ 9 callersFunctionmcp_token
( request: Request, server_id: str, user_api_key_dict: UserAPIKeyAuth = Depends(_mcp_o
litellm/proxy/management_endpoints/mcp_management_endpoints.py:2499
↓ 9 callersFunctionmessages_body
( model: str, request_nonce: str, *, safeguards: JsonValue | None = None, tools: Sequence[
tests/integration/_support/safeguards_rig.py:321
↓ 9 callersMethodmessages_body
(self, *, guardrails: tuple[str, ...] | None = None)
tests/integration/observability/test_presidio_streaming_output.py:186
↓ 9 callersFunctionmilvus_proxy_route
Enable using Milvus `/vectors` endpoint as a pass-through endpoint.
litellm/proxy/pass_through_endpoints/llm_passthrough_endpoints.py:943
↓ 9 callersFunctionmint_database_token
Mint a fresh database password for ``endpoint``, already percent-encoded.
litellm/proxy/db/token_auth.py:183
↓ 9 callersFunctionmint_internal_issuer_assertion
Signs one fresh, short-lived assertion; the caller must not cache the result, since a cached copy would defeat the point of re-minting on every ex
litellm/llms/base_llm/auth/internal_issuer.py:55
↓ 9 callersMethodmodel_dump
(self)
litellm/proxy/spend_tracking/spend_tracking_utils.py:409
↓ 9 callersMethodmodel_dump
(self)
litellm/llms/anthropic/chat/guardrail_translation/handler.py:151
↓ 9 callersFunctionmodel_for
`litellm_proxy/<group>`: every model call goes through the gateway.
tests/harness_e2e/conftest.py:43
↓ 9 callersFunctionn
(e,t,a)
litellm/proxy/_experimental/out/_next/static/chunks/3yrgf44a9g52e.js:7
↓ 9 callersFunctionnameless_rows
(count: int, first_index: int = 0)
tests/integration/_support/daily_activity.py:178
↓ 9 callersFunctionnative_call
(legacy: inspect.Signature, args: tuple[object, ...], kwargs: Mapping[str, object])
litellm/rust_bridge/public_call.py:102
↓ 9 callersFunctionnative_module
(py: Python<'_>)
litellm-rust/crates/python-bridge/src/lib.rs:105
↓ 9 callersMethodnearestDesc
(e,t=!1)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 9 callersFunctionnon_admin_caller
(scenario: Scenario, member: str, team: str, model: str)
tests/integration/authorization/_guardrail_opt_out.py:49
↓ 9 callersFunctiono
litellm/proxy/_experimental/out/_next/static/chunks/2j7hjtph8n74b.js:5
↓ 9 callersFunctiono
litellm/proxy/_experimental/out/_next/static/chunks/3yrgf44a9g52e.js:1
↓ 9 callersFunctionoK
(e,t,n,r)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 9 callersFunctionocr_cost
Args: model: str - model name custom_llm_provider: Optional[str] - custom LLM provider response: Optional[Any] - response
litellm/cost_calculator.py:2804
↓ 9 callersMethodof_unauthorized
( detail: str, *, www_authenticate: str | None = None, body: Mapping[str, str]
litellm/proxy/mcp_server/outbound_credentials/types.py:139
↓ 9 callersFunctionollama_pt
( model: str, messages: list )
litellm/litellm_core_utils/prompt_templates/factory.py:319
↓ 9 callersFunctionone_charged_row
(model_id: str)
tests/integration/spend/_background_responses.py:283
↓ 9 callersFunctiononly_request
(server: &MockServer)
litellm-rust/crates/inference-chat/tests/support/mod.rs:55
↓ 9 callersFunctionopen_messages
(core: RouterCore, **request: object)
tests/unit/test_router/core/test_parity_regress5_streams.py:58
↓ 9 callersFunctionopenai_error_event
(code: str, message: str)
tests/integration/_support/responses_stream.py:43
↓ 9 callersFunctionopenai_params
(wire: Wire)
tests/integration/routing/test_payment_required_mapping.py:73
↓ 9 callersMethodopenapi
(self)
tests/e2e/quota_management/spend_tracking/spend_e2e_client.py:689
↓ 9 callersFunctionoptional_mapping
(value: object)
litellm/rust_bridge/public_call.py:35
↓ 9 callersFunctionotel_audit_config
(audit_sinks: SpanSinks)
tests/integration/observability/conftest.py:28
↓ 9 callersFunctionotel_request_validation_exception_handler
(request: Request, exc: RequestValidationError)
litellm/proxy/proxy_server.py:2555
↓ 9 callersFunctionparse_fixture_mode
(raw: str)
tests/e2e/fixture_mode.py:46
↓ 9 callersFunctionparse_headers
Parse an OTLP ``"k=v,k=v"`` header string into a dict. ``OTEL_EXPORTER_OTLP_HEADERS`` is W3C Baggage encoded per the OTLP spec, so values are
litellm/integrations/otel/plumbing/providers.py:163
↓ 9 callersFunctionparse_http_body
(body: str)
litellm/proxy/mcp_server/tool_outcome.py:40
↓ 9 callersFunctionparse_tolerant_litellm_params
( litellm_params_data: Mapping[str, object], guardrail_name: str, params_model: type[_ParamsT] = L
litellm/proxy/guardrails/guardrail_registry.py:682
↓ 9 callersMethodpartial
(self)
tests/integration/management/test_team_delete_chaos.py:139
↓ 9 callersMethodpassword
()
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 9 callersMethodpatch_agent_in_db
Patch an agent in the database. Get the existing agent from the database and patch it with the new values. Args:
litellm/proxy/agent_endpoints/agent_registry.py:697
↓ 9 callersFunctionpdf_bytes
(pages: Sequence[Page])
tests/integration/_support/pdf_document.py:64
↓ 9 callersMethodpipeline
(self, transaction: bool = True)
litellm/caching/redis_cache.py:91
↓ 9 callersMethodplugins
()
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 9 callersFunctionpn
(e,t="Copied to clipboard")
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:10
↓ 9 callersFunctionpointer_for_key
(key: &str)
litellm-rust/crates/model-catalog/src/error.rs:250
↓ 9 callersFunctionpolicy_from_litellm_params
(litellm_params: Mapping[str, object])
litellm/proxy/guardrails/auto_router_compression.py:69
↓ 9 callersMethodpoll
(self)
tests/unit/proxy/client/cli/autoroute/test_process.py:28
↓ 9 callersMethodpoll_events_for_query
Poll until at least one matching event is searchable (the callback flushes in periodic batches and DataDog ingestion adds seconds of lag),
tests/e2e/logging/datadog_reader.py:168
↓ 9 callersFunctionpolled_tasks
(requests: tuple[Request, ...])
tests/integration/providers/_dashscope_video_support.py:84
↓ 9 callersMethodport
(&self)
litellm-rust/crates/redis-testing/src/tls.rs:89
↓ 9 callersFunctionpost
(gateway: Gateway, path: str, body: Mapping[str, JsonValue])
tests/integration/routing/test_payment_required_mapping.py:101
↓ 9 callersMethodpostgres_sqlstate
The SQLSTATE Postgres attached to a failed statement, as prisma surfaces it, or None.
litellm/proxy/db/exception_handler.py:387
↓ 9 callersMethodpre_call
(self, input: str | dict, api_key: str)
tests/unit/litellm_core_utils/test_realtime_streaming.py:4836
↓ 9 callersMethodprint_deployment
A copy of the deployment, of the type it was given, with its api key masked (``print_deployment``). A ``Deployment`` and its params read and w
litellm/router_rust/surface/members.py:148
↓ 9 callersMethodprobe
(self, name: str)
tests/integration/providers/_idle_upstream_support.py:678
↓ 9 callersMethodprocess_output_response
( self, response: object, guardrail_to_apply: "CustomGuardrail", litellm_loggi
litellm/llms/bedrock/passthrough/guardrail_translation/handler.py:1490
↓ 9 callersFunctionproxy_shutdown_event
(worker_heartbeat: ProxyWorkerHeartbeat | None = None)
litellm/proxy/proxy_server.py:1441
↓ 9 callersFunctionptu_identity_error
Why this config-declared reservation cannot be identified, else None. A deployment declared in config.yaml is otherwise keyed by a hash of its re
litellm/litellm_core_utils/ptu_pricing.py:127
↓ 9 callersMethodpublish
(self, channel: str, message: str)
tests/unit/proxy/auth/test_auth_checks.py:2259
↓ 9 callersMethodpubsub_numsub
(self, *channels: str)
tests/integration/_support/workers.py:34
↓ 9 callersMethodquery
(&self, read: impl Fn(&NameIndex) -> T)
litellm-rust/crates/model-catalog/src/store/mod.rs:87
↓ 9 callersFunctionquery_raw_within_statement_timeout
One raw read on the read target that Postgres cancels at ``timeout``. Prisma's HTTP client gives up on a statement after 30s anyway, so without
litellm/proxy/utils.py:5384
↓ 9 callersMethodquery_text
(query: str | Sequence[str])
litellm/llms/base_llm/vector_store/transformation.py:312
↓ 9 callersFunctionr
litellm/proxy/_experimental/out/_next/static/chunks/3yrgf44a9g52e.js:1
↓ 9 callersFunctionraise_public
Map a resolver CredError onto the proxy's public HTTP contract. The one edge that raises.
litellm/proxy/mcp_server/outbound_credentials/adapter.py:309
↓ 9 callersFunctionraise_token_exchange_challenge
Raise the RFC 9728 / RFC 6750 challenge an OBO (``token_exchange``) server returns when the caller's subject token is missing or the IdP rejected
litellm/proxy/mcp_server/outbound_credentials/adapter.py:379
↓ 9 callersFunctionread
(entry: Value)
litellm-rust/crates/model-catalog/src/wire/entry.rs:97
↓ 9 callersMethodread
( &self, key: K, load: impl Future<Output = Result<Option<V>, E>>, )
litellm-rust/crates/secrets-types/src/cache.rs:26
↓ 9 callersFunctionread_raw_json_body
(request: Request | None)
litellm/proxy/common_utils/http_parsing_utils.py:415
↓ 9 callersFunctionread_session_page
One page of sessions after ``session_cursor``, or after ``(page - 1) * page_size`` sessions, keeping only those whose outcome is ``status`` when g
litellm/proxy/spend_tracking/session_page.py:99
↓ 9 callersFunctionreadiness_for
(imported: float, workers: int)
tests/integration/_support/budget.py:29
↓ 9 callersFunctionreadiness_seconds
(workers: int = 1)
tests/integration/_support/budget.py:58
↓ 9 callersMethodreceive_text
(self)
tests/unit/llms/openai/live/test_handler.py:54
↓ 9 callersMethodreceived
(self, request_nonce: str)
tests/integration/messages_endpoint/test_anthropic_messages_caller_tools_wire.py:253
↓ 9 callersMethodrecord
(kwargs, completion_response, start_time, end_time)
tests/unit/test_router/test_router.py:7061
↓ 9 callersFunctionrecord_pre_routing_selection
Remember which model a pre-routing hook picked, so fallback lookup can key off it. Fallback resolution runs on an outer kwargs dict that ``*
litellm/router_utils/fallback_event_handlers.py:314
↓ 9 callersFunctionrecord_route_sweep
Append a passed or failed S2 report to its results ledger, when set.
tests/integration/security/_sweeps.py:481
↓ 9 callersFunctionrecord_to_dict
Project a database record into a mapping of column name to value.
litellm/repositories/base_repository.py:31
↓ 9 callersFunctionrecover_double_hashed_key_metadata
( prisma_client: PrismaClient, missing_keys: AbstractSet[str], )
litellm/proxy/spend_tracking/key_metadata_recovery.py:363
↓ 9 callersFunctionredact_credentials_in_payload
Return a copy of ``data`` where every value under a credential-named key is replaced by the shared ``REDACTED`` marker, nested mappings are recurs
litellm/litellm_core_utils/sensitive_data_masker.py:260
↓ 9 callersFunctionredact_message_input_output_from_logging
Removes messages, prompts, input, response from logging. This modifies the data in-place only redacts when litellm.turn_off_message_logging =
litellm/litellm_core_utils/redact_messages.py:759
↓ 9 callersMethodredact_sensitive_data
Redact sensitive information (tokens, secrets) from data before logging. Handles strings, dicts, and lists recursively. Keys contain
litellm/llms/databricks/common_utils.py:102
↓ 9 callersFunctionredis_cache_double
A `RedisCache` double for router tests. The native router stores connect from its settings, which name a closed local port, so they serve their me
tests/unit/conftest.py:27
↓ 9 callersMethodreject_config_owned_writes
Refuse a write to a setting the config file owns, rather than storing a value that never applies.
litellm/proxy/proxy_server.py:6192
↓ 9 callersFunctionrelease
(data: dict[str, object], user_api_key_dict: UserAPIKeyAuth, response: object)
tests/unit/proxy/pass_through_endpoints/test_pass_through_endpoints.py:8501
↓ 9 callersFunctionremote
(client: &NativeRedisClient)
litellm-rust/crates/python-bridge/src/redis/store/mod.rs:91
↓ 9 callersFunctionrender
(model: str, session: Session | None, config_dir: Path, use_color: bool, bar_width: int = BAR_WIDTH)
litellm/proxy/client/cli/commands/statusline_script.py:350
↓ 9 callersMethodreplicas_for
The replicas that serve `path`: every data-plane replica for an LLM route, and for a management route the control-plane replicas, since the da
tests/e2e/proxy_client.py:826
↓ 9 callersMethodreport
Record an outcome the router did not see itself, such as a health check. Returns whether the deployment was cooled down.
litellm-rust/crates/router-core/src/lib.rs:424
↓ 9 callersFunctionrepository_tag
(settings: ROISettings, repo: str)
litellm/proxy/roi_calculator/source.py:20
↓ 9 callersFunctionrequest_destinations
The destinations resolved for this request, empty outside a proxy request.
litellm/integrations/otel/plumbing/context.py:476
↓ 9 callersMethodreset
(self)
tests/integration/_support/device_login.py:270
← previousnext →8,201–8,300 of 202,479, ranked by callers