MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 5 callersFunctionprices
(rates: &[(Meter, Channel, ServiceTier, Band, f64)])
litellm-rust/crates/cost/src/price.rs:300
↓ 5 callersFunctionprint_verbose
(print_statement)
litellm/caching/caching.py:92
↓ 5 callersFunctionprintable
Labels come from the transcript, the proxy, and Claude Code's model cache, none of which this script controls, and every one is written to a termi
litellm/proxy/client/cli/commands/statusline_script.py:122
↓ 5 callersMethodpriority_header_fields
The request fields a priority header sets (was ``priority_header_update``).
litellm/router_api.py:1412
↓ 5 callersFunctionprisma_migration_lock
Hold only Prisma's key as an xact lock on a pinned connection with Prisma's session lock disabled. This keeps deploy independent of the coordinat
litellm-proxy-extras/litellm_proxy_extras/migration_lock.py:156
↓ 5 callersFunctionprobe_count_tokens
POST to `/v1/messages/count_tokens` for `model` and return the typed result. The Anthropic / LiteLLM `count_tokens` endpoint accepts a request bo
tests/e2e/claude_code/http_probe.py:110
↓ 5 callersFunctionprocess_callback
Process a single callback and return its data with environment variables
litellm/proxy/common_utils/callback_utils.py:697
↓ 5 callersMethodprocess_input_messages
Process input prompt by applying guardrails to text content. Args: data: Request data dictionary containing 'prompt' par
litellm/llms/openai/image_generation/guardrail_translation/handler.py:34
↓ 5 callersMethodprocess_model_name
(self, models: Iterable[Mapping[str, str]])
litellm/llms/gemini/common_utils.py:424
↓ 5 callersMethodprocess_output_response
Process output response by applying guardrails to completion text. Args: response: Text completion response object
litellm/llms/openai/completion/guardrail_translation/handler.py:137
↓ 5 callersMethodprocess_output_streaming_response
(self)
litellm/proxy/guardrails/guardrail_hooks/unified_guardrail/unified_guardrail.py:84
↓ 5 callersFunctionproject_info
Get information about a specific project Parameters: - project_id: *str* - The project id to fetch info for Example: ```bash
enterprise/litellm_enterprise/proxy/management_endpoints/project_endpoints.py:1072
↓ 5 callersFunctionproject_request
( bound: &Bound<'_, PyDict>, kwargs: &Bound<'_, PyDict>, )
litellm-rust/crates/python-bridge/src/routes/ocr/project.rs:120
↓ 5 callersFunctionprompt
(kind: str, index: int)
tests/integration/compatibility/test_codex_model_catalog.py:828
↓ 5 callersFunctionproviderForModel
(model: string)
ui/litellm-dashboard/src/app/(dashboard)/usage/_components/components/overview/modelProvider.ts:27
↓ 5 callersFunctionprovider_block
openai-completions is the one API shape every LiteLLM model serves. Real contextWindow/maxTokens matter: pi otherwise assumes 128k/16384, which
litellm/proxy/client/cli/commands/pi.py:218
↓ 5 callersMethodprovider_bodies
(self, marker: str)
tests/integration/observability/test_content_filter_logging_only.py:213
↓ 5 callersMethodprovider_hits
(self, marker: str)
tests/integration/observability/test_langtrace_delivery.py:239
↓ 5 callersMethodprovision_user
( self, *, marker: str, groups: tuple[str, ...], group_ids: tuple[str,
tests/e2e/idp.py:258
↓ 5 callersFunctionpublic_validation_errors
(errors: Sequence[ValidationErrorDetail])
litellm/proxy/common_utils/validation_error_body.py:12
↓ 5 callersFunctionpublish
The channel is sent as given, without the namespace, like redis-py's `publish`.
litellm-rust/crates/redis-client/src/commands.rs:394
↓ 5 callersFunctionpublish_plugin
( tx: PluginPublicationTransaction, publication: PluginPublication, expected: PluginPublicationBas
litellm/repositories/plugin_publication.py:122
↓ 5 callersFunctionpubsub_capable_client
(redis_cache: "RedisCache")
litellm/proxy/common_utils/config_sync_pubsub.py:99
↓ 5 callersMethodput_object
(self, Bucket, Key, Body, **_kwargs)
tests/unit/integrations/test_s3_v2.py:5672
↓ 5 callersFunctionq
(e)
litellm/proxy/_experimental/out/_next/static/chunks/39v0w3tgm6n4z.js:1
↓ 5 callersMethodquery
(self)
tests/integration/_support/aws_control_plane.py:45
↓ 5 callersMethodquery_embedding_executor
( embedding_executor: VectorStoreEmbeddingExecutor | None, router: Router | None, requ
litellm/llms/base_llm/vector_store/transformation.py:333
↓ 5 callersMethodqueue_spend_reset
(self, token: str, budget_reset_at: datetime | None, spend_decrement: float)
litellm/repositories/unit_of_work.py:36
↓ 5 callersFunctionr
(s,o,i,a)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 5 callersFunctionr
litellm/proxy/_experimental/out/_next/static/chunks/2j7hjtph8n74b.js:14
↓ 5 callersFunctionr
(e,r,s)
litellm/proxy/_experimental/out/_next/static/chunks/0o0w-zlajn03j.js:1
↓ 5 callersFunctionr
litellm/proxy/_experimental/out/_next/static/chunks/3i7pipwtkkn0_.js:1
↓ 5 callersFunctionr
(e)
litellm/proxy/_experimental/out/_next/static/chunks/0g3hfgh2z8the.js:1
↓ 5 callersFunctionr
(e,t=!0)
litellm/proxy/_experimental/out/_next/static/chunks/0dvgjd47fte1d.js:1
↓ 5 callersFunctionrC
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 5 callersFunctionrT
(e,t,n,r)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 5 callersFunctionraise_if_mock_testing_params_disallowed
Reject client-supplied mock testing params unless an admin opted in. Rejecting (rather than silently dropping) keeps a request that asked for a
litellm/proxy/route_llm_request.py:489
↓ 5 callersFunctionraise_if_unsafe_secret_name
Validate a secret name before it is used by a secret manager integration. Rejects ".." only as a path segment (bounded by "/" or the start/e
litellm/secret_managers/base_secret_manager.py:12
↓ 5 callersFunctionraise_if_unsupported_langfuse_version
Fail at logger construction rather than dropping every event at request time. v4 moved the callback onto OpenTelemetry, so on an older SDK the im
litellm/integrations/langfuse/langfuse.py:250
↓ 5 callersMethodraise_sensitive_data_route_exception
Raise an exception to reroute the request to a different model. Use this when sensitive data is detected and the guardrail is config
litellm/integrations/custom_guardrail.py:646
↓ 5 callersFunctionraise_user_oauth_challenge
Raise the 401 an ``authorization_code`` server returns at egress when the user has no token. Points at the server's RFC 9728 Protected Resource M
litellm/proxy/mcp_server/outbound_credentials/adapter.py:363
↓ 5 callersFunctionrank_chat_models_from_llm_provider
(custom_llm_provider: str)
litellm/litellm_core_utils/llm_request_utils.py:170
↓ 5 callersMethodrate_limits
(self, backfill_ttl: float | None = None)
litellm/redis/stores/connection.py:36
↓ 5 callersFunctionraw_deployment
(scenario: Scenario, model_name: str, litellm_params: Mapping[str, JsonValue])
tests/integration/providers/_coralbricks.py:373
↓ 5 callersFunctionrc
litellm/proxy/_experimental/out/_next/static/chunks/2k-fkubkt19cc.js:74
↓ 5 callersFunctionre
()
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 5 callersMethodread
(self)
litellm/litellm_core_utils/cli_keyring.py:166
↓ 5 callersMethodread
(names: BTreeSet<String>, lookup: &dyn Lookup)
litellm-rust/crates/gateway/src/redis/secrets.rs:15
↓ 5 callersFunctionreadAgentIdentity
(value: unknown)
ui/litellm-dashboard/src/app/(dashboard)/agents/_components/agent_identity.ts:31
↓ 5 callersFunctionread_batched_spend_counter
Bind-on-read for counters only known at read time (budget windows); the first reader pays the MGET.
litellm/proxy/spend_tracking/spend_counter_batch.py:279
↓ 5 callersFunctionread_branch_spend
( database: BranchSpendDatabase, start: date, end: date, repos: tuple[str, ...], *, casefold_repo: bool =
litellm/proxy/roi_calculator/branch_spend.py:16
↓ 5 callersFunctionread_failure
One status per failure kind, so a client can tell a bad cursor (400, fix the request) from a traversal it must restart (409), a result it cannot p
litellm/proxy/tracing_endpoints.py:141
↓ 5 callersFunctionread_gateway_user_emails
(prisma_client: _SpendPrismaClient)
litellm/proxy/roi_calculator/sync.py:191
↓ 5 callersFunctionread_pid_record
(path: Path | None = None)
litellm/proxy/client/cli/commands/autoroute/process.py:124
↓ 5 callersMethodread_resource
Fetch resource contents from the MCP server.
litellm/mcp_client/client.py:1542
↓ 5 callersMethodread_resource_from_server
Read resource contents from a specific MCP server.
litellm/proxy/mcp_server/mcp_server_manager.py:5047
↓ 5 callersFunctionread_return_to_cookie
The ``return_to`` this proxy sealed into ``litellm_cp_return_to``, or None when the cookie is missing, was not sealed here, or was tampered with.
litellm/proxy/management_endpoints/ui_sso.py:3248
↓ 5 callersFunctionread_rows_now
(key: str)
tests/integration/cost_calculation/conftest.py:123
↓ 5 callersFunctionreader
(locals: &Bound<'py, PyDict>, name: &str)
litellm-rust/crates/host-python/src/file_reader.rs:93
↓ 5 callersFunctionrebase
( value: JsonValue, offset_ns: int, namespace: str, response_pattern: re.Pattern[str], field: str = "" )
scripts/seed_tracing_fixtures.py:194
↓ 5 callersMethodreceive_text
(self)
litellm/llms/openai/live/handler.py:88
↓ 5 callersFunctionreceived
(server: &MockServer)
litellm-rust/crates/inference-ocr/tests/support/mod.rs:60
↓ 5 callersFunctionrecordUiEvent
(event: UiEvent)
ui/litellm-dashboard/src/lib/telemetry/uiEvents.ts:19
↓ 5 callersMethodrecord_authentication
(self, context: ManagedAgentContext)
litellm/proxy/agent_endpoints/identity_store.py:196
↓ 5 callersMethodrecord_operation_exception
( self, span_context: SpanContext, error_type: str, message: str, stac
litellm/integrations/otel/plumbing/events.py:35
↓ 5 callersFunctionrecord_output_text_chunk
Record an OUTPUT_TEXT_DONE chunk as a synthetic message item in ``text_only_items``. Real OUTPUT_ITEM_DONE events already captured in ``output
litellm/responses/sse_output_recovery.py:71
↓ 5 callersFunctionrecord_spend_counter_value
(counter_key: str, value: float)
litellm/proxy/spend_tracking/spend_counter_batch.py:265
↓ 5 callersFunctionrecord_upload
(root: Path, body: bytes, boundary: str)
tests/e2e_harness/test_provider_edge.py:389
↓ 5 callersFunctionrecord_upstream_responses
( spec: UpstreamEndpoint, case_input: InputT, sdk_call: Callable[[str, InputT], object], )
tests/rust-python-harness/shared/parity/fixtures/recording.py:223
↓ 5 callersFunctionredacted_model_call_details_for_custom_logger
Returns a copy of model_call_details with the request messages, instructions and streaming responses redacted when the custom logger has opte
litellm/litellm_core_utils/redact_messages.py:67
↓ 5 callersFunctionredacted_standard_logging_payload
Return a copy of a ``StandardLoggingPayload`` with its messages and response redacted. The success path redacts through ``perform_redaction`
litellm/litellm_core_utils/redact_messages.py:484
↓ 5 callersFunctionredis_breaker_state_counts
How many live Redis circuit breakers sit in each state, for the Prometheus services logger. Each Redis has one breaker, which every ``RedisCache`
litellm/caching/redis_cache.py:363
↓ 5 callersFunctionreduceStaticHeaders
(list: unknown)
ui/litellm-dashboard/src/app/(dashboard)/mcp-servers/_components/createServerPayload.ts:69
↓ 5 callersFunctionrefresh
()
tests/unit/proxy/mcp_server/outbound_credentials/test_redis_refresh_coordinator.py:64
↓ 5 callersFunctionrefresh
(source: impl CredentialSource, sender: watch::Sender<Latest>)
litellm-rust/crates/redis-connect/src/refresh.rs:186
↓ 5 callersMethodrefresh
Read router-core's indexes again when its registry moved since the last read.
litellm/router_rust/surface/indexes.py:377
↓ 5 callersMethodrefresh_catalog_authority
(auth: UserAPIKeyAuth | None)
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:1579
↓ 5 callersFunctionrefusal_events
(message: Mapping[str, JsonValue])
tests/integration/_support/anthropic_refusal.py:152
↓ 5 callersFunctionrefusal_stop_details
The ``stop_details`` object accompanying a translated ``stop_reason: "refusal"``.
litellm/llms/anthropic/messages/native/utils.py:47
↓ 5 callersFunctionrefused
(gateway: Gateway, wire: Wire, response: httpx.Response, message: str, hidden: str)
tests/integration/providers/test_vertex_credentials_source_wire.py:284
↓ 5 callersFunctionregister
(scenario: Scenario, body: Mapping[str, JsonValue])
tests/integration/management/test_claude_code_plugin_installation_preference.py:58
↓ 5 callersFunctionregister
(scenario: Scenario, prefix: str, response: StoredResponse)
tests/integration/compatibility/_status_code_audit.py:212
↓ 5 callersFunctionregister
(app, module)
tests/unit/proxy/test_route_priority.py:167
↓ 5 callersFunctionregisterAuthHeaderNameGetter
(getter: Getter<string>)
ui/litellm-dashboard/src/lib/http/runtime.ts:31
↓ 5 callersFunctionregisterAuthTokenGetter
(getter: Getter<string | null>)
ui/litellm-dashboard/src/lib/http/runtime.ts:35
↓ 5 callersFunctionregister_vertex
( scenario: Scenario, wire: Wire, token_url: str, *, extra_headers: JsonValue = None,
tests/integration/providers/_gemini_long_decode_support.py:154
↓ 5 callersFunctionregistry_key_for_prompt
(prompt: PromptSpec)
litellm/proxy/prompts/prompt_registry.py:73
↓ 5 callersFunctionreject_api_key_secret_reference
(params: Mapping[str, object])
litellm/litellm_core_utils/credential_accessor.py:55
↓ 5 callersFunctionreject_legacy_identity
(params: Mapping[str, object] | None)
litellm/proxy/agent_endpoints/identity.py:15
↓ 5 callersFunctionreject_unknown_gemini_tts_voice_keys
( model: str, voice: Mapping[str, object], drop_params: bool, llm_provider: str | None )
litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py:469
↓ 5 callersFunctionrejected_token_count
(rejection: _Rejection, litellm_call_id: str, request_model: str | None)
litellm/proxy/common_utils/token_count_log.py:83
↓ 5 callersFunctionrelay
uvicorn never half-closes a WebSocket, so Python's EOF means the socket is gone and the relay ends. Only a client that is gone ends it from the other
litellm-rust/crates/gateway-forward/src/websocket.rs:58
↓ 5 callersMethodrelease
(self, size: int)
litellm/proxy/lens/buffering.py:29
↓ 5 callersFunctionrelease_budget_reservation_on_cancel
Reconcile a still-open reservation when the request is cancelled mid-flight. A client disconnect or timeout cancels the request task, which surfa
litellm/proxy/spend_tracking/budget_reservation.py:399
↓ 5 callersFunctionrelease_in_background
(release: Coroutine[object, object, None])
litellm/litellm_core_utils/attempt_settlement.py:73
↓ 5 callersFunctionreleased
Wait for ``releases`` without ever cancelling them: a cancelled waiter leaves them running.
litellm/litellm_core_utils/attempt_settlement.py:80
↓ 5 callersFunctionreload_serving_verdict
Judge a write-triggered reload by diffing the router's serving state instead of trusting any layer of the reload stack to report its own failure.
litellm/proxy/management_endpoints/model_management_endpoints.py:3939
↓ 5 callersMethodremove
(&mut self, key: &str)
litellm-rust/crates/redis-store/src/dual_cache/batch_read.rs:237
↓ 5 callersFunctionremoveAt
(items: readonly T[], index: number)
ui/litellm-dashboard/src/app/(dashboard)/playground/components/systemOneUI/lib/formPayload.ts:61
↓ 5 callersMethodremoveObserver
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
← previousnext →16,401–16,500 of 202,479, ranked by callers