MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 6 callersFunctionr1
(e)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 6 callersFunctionrN
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 6 callersFunctionrR
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 6 callersFunctionraise_denied_scoped_mcp_access
A scoped request (``/mcp/<name>`` path or ``x-mcp-servers`` header) resolved to zero allowed servers, so the denial must be loud: a silent 200 wit
litellm/proxy/mcp_server/operations.py:716
↓ 6 callersMethodraise_if_lens_rename_pending
(connect: LensCheckConnect | None = None)
litellm-proxy-extras/litellm_proxy_extras/utils.py:646
↓ 6 callersFunctionraise_if_not_oauth2
Reject a server without upstream OAuth from the gateway's authorize/token/register flow. The client-forwarded token modes (``true_passthrough`` /
litellm/proxy/mcp_server/discoverable_endpoints.py:756
↓ 6 callersMethodraise_rate_limit_error
Raise a 429 with a retry-after header for litellm-proxy parallel-request limits. Always raises :class:`ProxyRateLimitError` — never
litellm/proxy/hooks/parallel_request_limiter.py:188
↓ 6 callersMethodread
(s)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 6 callersFunctionread_list
Reads a list field from a dict or an object; anything else reads as empty. The entries are the reply's own objects, so writing through them edits
litellm/proxy/guardrails/guardrail_hooks/llm_shield_proxy/payload.py:103
↓ 6 callersFunctionread_of
(response: httpx.Response)
tests/integration/spend/_background_responses.py:199
↓ 6 callersFunctionread_reload_schedule
(prisma_client: PrismaClient, param_name: str)
litellm/proxy/common_utils/periodic_reload_schedule.py:174
↓ 6 callersFunctionread_scope_sql
(scope: ReadScope, next_param: int)
litellm/proxy/spend_tracking/spend_management_endpoints.py:4858
↓ 6 callersMethodreader
(self)
litellm/proxy/tracing_endpoints.py:75
↓ 6 callersFunctionrealtime_model
Return the provisioned deployment name for this provider. Every provider in PROVIDERS is provisioned at session start, so a missing entry is a har
tests/e2e/llm_translation/realtime/realtime_client.py:116
↓ 6 callersFunctionrealtime_request_body
Generate the realtime websocket request body. Cached with LRU semantics to avoid repeated string formatting work while keeping memory usage b
litellm/proxy/common_utils/realtime_utils.py:9
↓ 6 callersFunctionrealtime_ssl_for_url
(url: str)
litellm/llms/custom_httpx/http_handler.py:497
↓ 6 callersFunctionreapply_runtime_model_cost_registrations
Re-apply runtime model metadata on top of a freshly adopted cost map. Adopting a new catalog replaces ``litellm.model_cost`` wholesale, which on
litellm/utils.py:4355
↓ 6 callersFunctionreasoning_text
(deltas: Sequence[Mapping[str, JsonValue]])
tests/integration/_support/anthropic_thinking.py:196
↓ 6 callersFunctionreceive
()
litellm/proxy/management_endpoints/policy_endpoints/endpoints.py:241
↓ 6 callersFunctionreconstruct_model_name
Reconstruct full model name with provider prefix for logging.
litellm/litellm_core_utils/core_helpers.py:643
↓ 6 callersMethodrecord
(&self, op: Seen)
litellm-rust/crates/inference-messages/tests/messages/stream.rs:66
↓ 6 callersFunctionrecord_auth_resolution
(server_id: str, source: AuthResolution)
litellm/proxy/mcp_server/mcp_debug.py:139
↓ 6 callersMethodrecord_hit
(self)
litellm/proxy/db/db_lookup_gate.py:44
↓ 6 callersMethodrecord_key
( self, mount: str, outcome: str, test_key: str, method: str, url: str, headers: Mapping[str,
tests/e2e/provider_cache.py:544
↓ 6 callersFunctionredact_credential_headers
Return a copy of `headers` with credential-bearing values masked. `clean_headers` deliberately preserves some credential headers so they can be
litellm/proxy/litellm_pre_call_utils.py:1628
↓ 6 callersFunctionredact_secrets
Public API: redact known secret/credential patterns from an arbitrary string. Use this for code paths that bypass the logging system — e.g. Slack
litellm/_logging.py:120
↓ 6 callersFunctionrefresh_user_oauth_token
Attempt to refresh a per-user OAuth2 token using its stored refresh_token. POSTs to ``server.effective_token_url`` with ``grant_type=refresh_toke
litellm/proxy/mcp_server/db.py:2053
↓ 6 callersFunctionrefusal_delta
(stop_details: Mapping[str, object] | None = None)
tests/unit/test_router/core/streaming_helpers.py:266
↓ 6 callersFunctionrefusal_fields
()
tests/integration/messages_endpoint/safeguards/test_safeguard_results_synthesized_streams.py:193
↓ 6 callersFunctionrefusing_forward_proxy
Owned HTTP forward proxy: records the host each CONNECT asks for and refuses the tunnel with 403. A litellm proxy booted with ``HTTPS_PROXY`` poi
tests/integration/_support/forward_proxy.py:35
↓ 6 callersFunctionregister
(scenario: Scenario, wire: Wire, credentials: JsonValue)
tests/integration/providers/test_vertex_credentials_source_wire.py:74
↓ 6 callersMethodregister
(app, module)
tests/unit/proxy/test_proxy_server_endpoints_and_startup.py:10293
↓ 6 callersFunctionregistry
(entries: Value)
litellm-rust/crates/router-core/src/resolve.rs:449
↓ 6 callersMethodregistry
(self)
litellm/router_rust/invoke/endpoints.py:140
↓ 6 callersMethodreject
(s,o)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 6 callersMethodrejected_writes
(self, incoming: Mapping[str, JsonValue])
litellm/proxy/config_resolvers/settings_store.py:69
↓ 6 callersFunctionrelaxed
(server: RecordingServer)
tests/test_litellm_rust/support/recording_server.py:130
↓ 6 callersMethodrelease
(self)
litellm/proxy/lens/buffering.py:43
↓ 6 callersMethodrelease
(self)
tests/integration/_support/database_relay.py:106
↓ 6 callersMethodrelease
(self, key: str, lease: CaptureLease)
tests/e2e/provider_cache.py:125
↓ 6 callersFunctionreload_active_user_by_id
``None`` when the user is live, else the precise failure ``load_active_user_by_id`` found.
litellm/proxy/mcp_server/bridge_token_flow.py:312
↓ 6 callersMethodreload_search_tools_from_db
Refresh this worker's router from the search tools table. Driven by the management endpoints so the worker that served the write is correct
litellm/proxy/proxy_server.py:9901
↓ 6 callersMethodremove_policy
Remove a policy by name. If a config-defined policy shares the name, it is restored immediately instead of waiting for the next DB sy
litellm/proxy/policy_engine/policy_registry.py:366
↓ 6 callersMethodremove_unregistered_routes
(app: FastAPI)
litellm/proxy/pass_through_endpoints/pass_through_endpoints.py:3762
↓ 6 callersFunctionrender
(val: number, lng: string)
ui/litellm-dashboard/src/i18n/formats.test.ts:98
↓ 6 callersFunctionrender_trace_results
(results: Sequence[CaseResult])
tests/rust-python-harness/strategies/trace_parity/reporting.py:178
↓ 6 callersFunctionreplace_file_search_tools
Replace all file_search tools with a single function tool. Returns: (new_tools_list, all_vector_store_ids)
litellm/responses/file_search/emulated_handler.py:108
↓ 6 callersFunctionrequest
(call_id: str)
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:5549
↓ 6 callersFunctionrequest_caller_aliases
(user_api_key_dict: UserAPIKeyAuth, listed_team_id: str | None = None)
litellm/proxy/proxy_server.py:12665
↓ 6 callersFunctionrequest_contains_image_content
Inspect native message content without translating or validating the request.
litellm/litellm_core_utils/prompt_templates/common_utils.py:557
↓ 6 callersFunctionrequest_from
(value: serde_json::Value)
litellm-rust/crates/llms/src/azure_ai/messages/transformation.rs:144
↓ 6 callersFunctionrequest_messages
(request: Request)
tests/integration/providers/_tool_search_support.py:147
↓ 6 callersMethodrequest_reply
( &self, request: impl FnOnce(Reply<A>) -> HostRequest<P> + Send, )
litellm-rust/crates/host/src/machine/context.rs:43
↓ 6 callersFunctionrequest_skips_budget_checks
( route: str, model: str | list[str] | None, llm_router: Router | None, *, request_body: M
litellm/proxy/auth/auth_checks.py:1123
↓ 6 callersFunctionrequest_spend_log_flush
Wake this client's queue monitor now rather than leaving the rows for its next poll. The Responses API hands the client an id it can chain from s
litellm/proxy/utils.py:8632
↓ 6 callersFunctionrequiredRule
(message: string)
ui/litellm-dashboard/src/app/(dashboard)/guardrails/_components/GuardrailFormField.tsx:48
↓ 6 callersFunctionresets_on
(candidates: tuple[Gateway, ...], key: str, model: str)
tests/integration/management/test_budget_reset_windows.py:206
↓ 6 callersMethodresolution
(self)
litellm/proxy/mcp_server/mcp_debug.py:157
↓ 6 callersFunctionresolve
(plugin: object)
litellm/proxy/proxy_server.py:20755
↓ 6 callersFunctionresolve
(route)
tests/unit/proxy/auth/test_handle_jwt.py:5838
↓ 6 callersFunctionresolve
(headers: Headers, auth: AuthScheme)
litellm-rust/crates/llms/src/base_llm/auth.rs:133
↓ 6 callersMethodresolve
Request timeout resolution order (first non-None wins): 1. ``model_timeout`` (call argument / merged ``litellm_params``) 2.
litellm/litellm_core_utils/completion_timeout.py:123
↓ 6 callersMethodresolve
(include_file: str, declared_in: str)
litellm/proxy/proxy_server.py:6083
↓ 6 callersMethodresolve
Build the context from a gateway handler's kwargs. ``user_api_key_auth`` is read from both metadata keys because routes differ:
litellm/responses/mcp/request_context.py:45
↓ 6 callersMethodresolve
( self, auth: &AwsAuthService, env_lookup: &(dyn Fn(&str) -> Option<String> + Sync),
litellm-rust/crates/auth-aws/src/aws.rs:771
↓ 6 callersMethodresolve
(mut self)
litellm-rust/crates/redis-client/src/batch.rs:121
↓ 6 callersFunctionresolve_api_base
(api_base: str | None)
litellm/llms/edenai/common_utils.py:32
↓ 6 callersMethodresolve_api_key
(api_key: str | None)
litellm/llms/databricks/common_utils.py:188
↓ 6 callersFunctionresolve_base_ref
(base_ref: str | None, repo_root: Path)
scripts/default_branch.py:43
↓ 6 callersFunctionresolve_bedrock_region
( model_region: Option<&str>, params: &AwsParams, env_lookup: &dyn Fn(&str) -> Option<String>, )
litellm-rust/crates/auth-aws/src/aws.rs:718
↓ 6 callersMethodresolve_classified_tier
Resolve a classifier reply to the active tier it names, or None when it names none.
litellm/router_strategy/complexity_router/config.py:2074
↓ 6 callersFunctionresolve_langfuse_credentials
( langfuse_public_key=None, langfuse_secret=None, langfuse_secret_key=None, langfuse_host=None
litellm/integrations/langfuse/langfuse.py:297
↓ 6 callersMethodresolve_metadata_value
(request_data: Mapping[str, object] | None, key: str)
litellm/proxy/guardrails/guardrail_hooks/akto/akto.py:335
↓ 6 callersFunctionresolve_model_latest_alias
(model: str, provider: str)
litellm/litellm_core_utils/model_latest_aliases.py:56
↓ 6 callersMethodresolve_model_name_from_model_id
( self, model_id: str | None, custom_llm_provider: str | None = None )
litellm/router_deprecated.py:536
↓ 6 callersFunctionresolve_oci_schema_anyof
Resolve Pydantic v2 ``Optional[T]`` → ``anyOf`` patterns. Pydantic v2 emits ``{"anyOf": [{"type": "T"}, {"type": "null"}]}`` for ``Optional[T
litellm/llms/oci/common_utils.py:592
↓ 6 callersFunctionresolve_oci_schema_refs
Inline all ``$ref``/``$defs`` references — OCI does not support JSON Schema ``$ref``.
litellm/llms/oci/common_utils.py:561
↓ 6 callersFunctionresolve_openapi_tool_auth
The caller's upstream credential for one ``spec_path`` server, for both OpenAPI dispatch arms. A per-server ``x-mcp-{alias}-authorization`` wins
litellm/proxy/mcp_server/mcp_server_manager.py:1416
↓ 6 callersFunctionresolve_otlp_http_tls
(signal: Literal["TRACES", "METRICS", "LOGS"])
litellm/integrations/otel/plumbing/otlp_tls.py:28
↓ 6 callersFunctionresolve_prediction_target
(params: LiteLLM_Params)
litellm/llms/anthropic/prompt_cache_prediction.py:632
↓ 6 callersMethodresolve_public_name
Resolve a public team name back to the internal routing key the router indexes by, so `GET /v1/models/{id}` accepts the name the listing retur
litellm/proxy/common_utils/model_listing_utils.py:502
↓ 6 callersFunctionresolve_routing_plugins
Resolves a list of routing-plugin entries to live `RoutingPlugin` instances. Each string entry is resolved through `get_instance_fn` (the sam
litellm/proxy/proxy_server.py:5436
↓ 6 callersMethodresolve_streaming_flag
Streaming flag resolution order (later wins): default < guardrail attribute < guardrail_config dict < this callback's optional_params.
litellm/proxy/guardrails/guardrail_hooks/unified_guardrail/unified_guardrail.py:1078
↓ 6 callersFunctionresolve_structured_messages
Normalize a request's messages to OpenAI-spec chat-completions shape, regardless of which API surface produced them (chat completions, An
litellm/litellm_core_utils/prompt_templates/factory.py:6588
↓ 6 callersFunctionresolved
a host hands its credential_provider object over as a handle, so these tests do the same
litellm-rust/crates/redis-connect/tests/connect.rs:15
↓ 6 callersFunctionresourceLiteLLMProject
()
terraform/provider/litellm/resource_project.go:46
↓ 6 callersFunctionresourceLiteLLMTeamUpdate
(d *schema.ResourceData, m interface{})
terraform/provider/litellm/resource_team.go:317
↓ 6 callersFunctionresources
()
litellm-rust/crates/inference-testing/src/lib.rs:19
↓ 6 callersMethodresponse_api_handler
( self, previous_response_client_ids: tuple[str, ...] = (), /, *, mode
litellm/responses/litellm_completion_transformation/handler.py:76
↓ 6 callersFunctionresponse_header
(head: str, name: str)
tests/e2e_harness/test_provider_edge.py:229
↓ 6 callersFunctionresponses_body
(model: str, marker: str)
tests/integration/providers/_cache_control_marks_support.py:168
↓ 6 callersFunctionresponses_peer
(cert: tuple[Path, Path])
tests/integration/providers/test_responses_websocket_session_limit.py:376
↓ 6 callersMethodresponses_raw
POST /v1/responses (OpenAI Responses API): raw outcome judged by status/body/headers, for tests that need x-litellm-call-id. max_outpu
tests/e2e/logging/logging_client.py:547
↓ 6 callersFunctionresponses_reasoning_items_from_thinking_blocks
Build Responses API `reasoning` input items from Anthropic thinking blocks. A block carrying encrypted reasoning replays the item it came from by
litellm/litellm_core_utils/prompt_templates/reasoning_utils.py:327
↓ 6 callersFunctionrestamp_anthropic_stream_chunk_model
Return ``chunk`` with the ``message_start`` model replaced by ``requested_model``. Chunks that carry no model are returned unchanged.
litellm/proxy/anthropic_endpoints/streaming_model_restamp.py:64
↓ 6 callersFunctionresume_unsettled_background_interactions
Pick up every settlement no process has finished, which is what a replica that died mid-poll or mid-settlement leaves behind. A row whose cla
litellm/interactions/background_cost_polling.py:872
↓ 6 callersFunctionretried_failure
`KeyError`'s classes, most specific first. The failure a retry loop decides by: an attempt's, or the exception a failed pick raised (Python's loop ret
litellm-rust/crates/router-core/src/call.rs:279
↓ 6 callersFunctionretrieve_batch
( client: BatchClient, batch_id: str, *, key: str, provider: str | None )
tests/e2e/batches/test_batches_e2e.py:134
↓ 6 callersFunctionrewrite
(setup_model: str)
litellm/proxy/pass_through_endpoints/llm_passthrough_endpoints.py:4195
↓ 6 callersFunctionrewrite_query_ids
Walk query param values and resolve any passthrough managed IDs. Returns *params* unchanged (same object) when nothing is resolved.
litellm/proxy/pass_through_endpoints/managed_id_rewriter.py:1262
← previousnext →13,401–13,500 of 202,479, ranked by callers