MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 10 callersMethodnear
(e,t=1)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 10 callersFunctionnonce
()
tests/integration/messages_endpoint/test_anthropic_messages_caller_tools_wire.py:84
↓ 10 callersFunctionnormalize_nonempty_secret_str
Strip whitespace and treat None, '', and whitespace-only strings as unset. Use when pairing secrets (mutual exclusion, optional auth) so whi
litellm/secret_managers/main.py:323
↓ 10 callersMethodnotify
()
litellm/proxy/_experimental/out/_next/static/chunks/3lb5gjzjxjxem.js:1
↓ 10 callersFunctiono
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/14v3irtjifjt2.js:1
↓ 10 callersFunctionoauth_endpoints_unresolved
Whether endpoint or eligible client metadata discovery is still pending. The flow comes from ``effective_oauth2_flow``, the one column-first, sha
litellm/proxy/mcp_server/mcp_server_manager.py:807
↓ 10 callersFunctionobject_mapping
``value`` as a str-keyed mapping, None for anything else. Vendor JSON arrives untyped, so this is the one place that turns it into something the r
litellm/litellm_core_utils/llm_cost_calc/guardrail_cost.py:112
↓ 10 callersMethodobjects
(self)
tests/integration/observability/test_batch_line_items_cold_storage.py:92
↓ 10 callersMethodonly_body
(self, path: str = "/v1/responses")
tests/integration/providers/test_openai_responses_tool_search_wire.py:156
↓ 10 callersFunctiononly_request
(wire: Wire)
tests/integration/providers/_sagemaker_chat_container.py:151
↓ 10 callersMethodoperation
(self)
litellm/proxy/mcp_server/catalog.py:196
↓ 10 callersFunctionoperator_sink_scopes
The accounts the operator's own exporters write to, in destination terms. Every v2 logger's config counts, since each logger exports through its
litellm/integrations/otel/plumbing/providers.py:1707
↓ 10 callersFunctionoptionally_handle_anthropic_oauth
Handle Anthropic OAuth token detection and header setup. If an OAuth token is detected in the Authorization header (any casing), extract
litellm/llms/anthropic/common_utils.py:561
↓ 10 callersFunctionpage
(name: str, cursor: str | None = None, revision: str = "stable")
tests/unit/proxy/mcp_server/test_catalog.py:136
↓ 10 callersFunctionparse
(variables: &[(&str, &str)])
litellm-rust/crates/gateway-forward/tests/settings.rs:14
↓ 10 callersFunctionparse_anthropic_error_event
Extract ``(error_type, message, http_status_code)`` from an Anthropic SSE ``event: error`` chunk (raw bytes or an already-decoded dict), or N
litellm/llms/anthropic/messages/native/streaming_iterator.py:160
↓ 10 callersFunctionparse_model_quota_setting
Any falsy value is off and any other value requires both limits, as the boolean setting always behaved, so only a mapping that names a policy key
litellm/proxy/management_helpers/model_quota_policy.py:69
↓ 10 callersFunctionparse_team_metadata_schema
Parse ``general_settings.team_metadata_schema``; raises on a malformed schema so config load fails fast.
litellm/proxy/management_helpers/team_metadata_validation.py:79
↓ 10 callersMethodpatterns
(self)
litellm/router_rust/surface/indexes.py:275
↓ 10 callersFunctionpayloads
(text: str, stream: bool)
tests/integration/providers/_gemini_long_decode_support.py:302
↓ 10 callersFunctionpersist_deleted_verification_tokens
Persist deleted verification token records by transforming and saving them.
litellm/proxy/management_endpoints/key_management_endpoints.py:5882
↓ 10 callersMethodpersonal_key_agent_billing_auth
( auth: UserAPIKeyAuth, agent_id: str, *, strict: bool = False, )
litellm/proxy/agent_endpoints/auth/agent_permission_handler.py:440
↓ 10 callersMethodpick_model
(listed)
tests/unit/proxy/client/cli/test_configure_commands.py:704
↓ 10 callersMethodpolicy_applies
Predicate telling whether a policy exists and any policy in its inheritance chain applies to the context. Admissions where the
litellm/proxy/policy_engine/policy_matcher.py:136
↓ 10 callersMethodpre_call_hook
( self, user_id: str | None, messages: list[AllMessageValues], function_call:
litellm/integrations/gitlab/gitlab_prompt_manager.py:356
↓ 10 callersMethodprepare_request_auth
Preview the authenticated request without sending it, closing the auth flow afterwards.
litellm/mcp_client/client.py:591
↓ 10 callersMethodprepare_token_exchange_parameters
Prepare token exchange parameters for Generic SSO. Args: request: Request object generic_include_client_id:
litellm/proxy/management_endpoints/ui_sso.py:4141
↓ 10 callersFunctionpresent
(request: NativeCall, name: str)
litellm/rust_bridge/detectors.py:30
↓ 10 callersFunctionprint_startup_warning
Write a boot-time warning as print_startup_message writes a message, even with suppress_startup_output on, since it names something the operator h
litellm/_logging.py:978
↓ 10 callersMethodprocess
(self, chunk: object)
litellm/proxy/anthropic_endpoints/streaming_model_restamp.py:114
↓ 10 callersMethodprocess_input_messages
Process input text by applying guardrails. Args: data: Request data dictionary containing 'input' parameter
litellm/llms/openai/speech/guardrail_translation/handler.py:33
↓ 10 callersFunctionq
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 10 callersFunctionqualify_agentic_followup_model
Resolve the model an agentic follow-up re-dispatches, keeping a hook's own provider.
litellm/litellm_core_utils/core_helpers.py:74
↓ 10 callersFunctionready
An op answered without Redis, from memory.
litellm-rust/crates/redis-store/src/request.rs:217
↓ 10 callersFunctionreceived_text
(request: Request)
tests/integration/providers/_gemini_tts_voice_support.py:115
↓ 10 callersMethodrecord_native_bedrock_guardrail
Bill the guardrail trace a Bedrock response carries when the deployment's ``guardrailConfig`` ran it, by recording it as guardrail information
litellm/litellm_core_utils/litellm_logging.py:3070
↓ 10 callersFunctionrecovered_cost
(gateway: Gateway, scenario: Scenario, model: str)
tests/integration/spend/_partial_stream.py:163
↓ 10 callersMethodredactSensitiveData
redactSensitiveData masks sensitive information in logs
terraform/provider/litellm/client.go:539
↓ 10 callersFunctionredact_string
Scrub known secret/credential patterns from *value* and return the result.
litellm/litellm_core_utils/secret_redaction.py:98
↓ 10 callersFunctionredis_sync
Runs one Redis operation with the GIL released, for the router's sync paths.
litellm-rust/crates/python-bridge/src/redis/store/mod.rs:63
↓ 10 callersFunctionredis_text
(redis: Redis, key: str)
tests/integration/management/test_budget_reset_windows.py:175
↓ 10 callersFunctionreference
(session: str)
tests/integration/routing/_prompt_cache_pin_support.py:59
↓ 10 callersMethodrefresh
Read router-core's list again after a write (only what was appended, when that is all that changed). The index views are read again when an en
litellm/router_rust/surface/registry.py:1371
↓ 10 callersFunctionrefresh_cached_team
Refresh the cached team object after a DB write and broadcast invalidations to other instances. Every endpoint that mutates `litellm_teamtab
litellm/proxy/management_endpoints/team_endpoints.py:591
↓ 10 callersMethodrefund
( self, reservation: BatchEnqueuedTokenReservation, litellm_parent_otel_span: "Span |
litellm/proxy/hooks/batch_enqueued_tokens.py:334
↓ 10 callersFunctionregenerate_key_fn
Regenerate an existing API key while optionally updating its parameters. Parameters: - key: str (path parameter) - The key to regenerate
litellm/proxy/management_endpoints/key_management_endpoints.py:6485
↓ 10 callersFunctionregister_datadog_mcp
Register the core Datadog toolset with its credentials from the env. By default the server exposes only `search_datadog_logs`; pass `allowed_tools
tests/e2e/mcp/datadog_mcp.py:41
↓ 10 callersFunctionregister_guardrail
Register a guardrail for onboarding (team submission). Accepts a guardrail config in the [Generic Guardrail API](https://docs.litellm.ai
litellm/proxy/guardrails/guardrail_endpoints.py:756
↓ 10 callersFunctionregister_priced_model
Register a deployment with explicit custom rates (deleted on teardown) and return its unique model name.
tests/e2e/quota_management/spend_tracking/cost_rows.py:196
↓ 10 callersMethodregistry
(self)
tests/unit/proxy/db/test_model_access_group_spend.py:23
↓ 10 callersFunctionreject_server_owned_wif_params
Raise ``ValueError`` if a mapping that did not come from deployment config carries a server-owned workload identity federation or OAuth token-exch
litellm/types/router.py:1553
↓ 10 callersMethodrelease_lock
(self, cronjob_id: str)
tests/unit/proxy/db/test_gateway_request_tracking.py:508
↓ 10 callersMethodreload_admitted_user
Reload the live user an interactively-minted envelope references and admit them as themselves. The user's own object permission and ``org_id`
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:1460
↓ 10 callersFunctionrender_custom_prompt_template
( template: Mapping[str, object], messages: Sequence[Mapping[str, object]], model: str, llm_pr
litellm/litellm_core_utils/prompt_templates/factory.py:6122
↓ 10 callersMethodrender_template
(self, template_id: str, variables: Mapping[str, object] | None = None)
litellm/integrations/gitlab/gitlab_prompt_manager.py:228
↓ 10 callersFunctionrender_token_fault
RFC 6749 §5.2 response for a token-endpoint fault. Caller-actionable rejections relay the upstream's code on the status that code implies (401 for
litellm/proxy/mcp_server/faults/render_oauth.py:56
↓ 10 callersFunctionrendered_tokens
(pages: Sequence[Page])
tests/integration/_support/pdf_document.py:34
↓ 10 callersMethodreplica
(self)
tests/unit/proxy/hooks/test_model_max_budget_limiter.py:449
↓ 10 callersFunctionreply_text
(status: int, raw: bytes, *, streaming: bool)
tests/integration/messages_endpoint/test_anthropic_messages_caller_tools_wire.py:284
↓ 10 callersMethodrequest_capacity
Charge one non-generation provider request to RPM and hold its concurrency slot.
litellm/proxy/hooks/parallel_request_limiter_v3.py:3779
↓ 10 callersFunctionrequest_tags_from_metadata
Tags for the batch-cost spend row: the request's own tags when it sent any, otherwise the key's tags, which auth exposes as user_api_key_auth_meta
litellm/proxy/pass_through_endpoints/llm_provider_handlers/batch_attribution.py:60
↓ 10 callersMethodreserve
( self, tokens: int, scopes: tuple[BatchEnqueuedTokenScope, ...], litellm_pare
litellm/proxy/hooks/batch_enqueued_tokens.py:229
↓ 10 callersFunctionreserve_lazy_slot
Record the route the feature's router used to be included after, so its routes are spliced back in there once it loads and keep the same precedenc
litellm/proxy/_lazy_features.py:414
↓ 10 callersFunctionresetToDataState
()
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 10 callersFunctionreset_at
(started_at: float, seconds: float)
tests/integration/management/test_budget_reset_windows.py:171
↓ 10 callersMethodreset_budget
Gets all the non-expired keys for a db, which need spend to be reset Resets their spend Updates db Runs on one pod
litellm/proxy/common_utils/reset_budget_job.py:564
↓ 10 callersMethodreset_budget_for_litellm_users
Resets the budget for all LiteLLM Internal Users if their budget has expired
litellm/proxy/common_utils/reset_budget_job.py:1240
↓ 10 callersFunctionreset_key_spend_fn
( key: str, data: ResetSpendRequest, user_api_key_dict: UserAPIKeyAuth = Depends(user_api_key_auth
litellm/proxy/management_endpoints/key_management_endpoints.py:6929
↓ 10 callersFunctionresolve
()
ui/litellm-dashboard/src/app/(dashboard)/model-insights/_components/ModelInsightsView.test.tsx:94
↓ 10 callersMethodresolve
(s,o)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 10 callersFunctionresolve_agent_access_group_ceiling
``None`` when the agent has no access groups attached, so nothing is capped. A named group that provably does not exist is dropped from the ceilin
litellm/proxy/agent_endpoints/auth/agent_access_groups.py:74
↓ 10 callersFunctionresolve_client_ip
Resolve the real client IP, trusting X-Forwarded-For only when the direct peer is itself a configured trusted proxy. Walks the header right-to-lef
litellm/proxy/auth/network.py:79
↓ 10 callersFunctionresolve_complexity_router_plugins
Resolves `complexity_router_config["plugins"]` dotted-path strings to live instances in place, via `resolve_routing_plugins`, and `comple
litellm/proxy/proxy_server.py:5474
↓ 10 callersFunctionresolve_database_token_auth
Resolve the token strategy from the environment, raising when both toggles are set.
litellm/proxy/db/token_auth.py:276
↓ 10 callersFunctionresolve_health_check_mode
Effective mode for a deployment's health-check probe. Prefers operator-set `model_info.mode`; then the mode the provider requires for th
litellm/proxy/health_check.py:93
↓ 10 callersMethodresolve_key_team_agent_access
( user_api_key_auth: UserAPIKeyAuth | None, *, strict: bool = False, lookup_ke
litellm/proxy/agent_endpoints/auth/agent_permission_handler.py:168
↓ 10 callersFunctionresolve_oci_credentials
Merge OCI credentials from optional_params (explicit, always wins) and environment variables (fallback). Returns a dict with resolved va
litellm/llms/oci/common_utils.py:163
↓ 10 callersFunctionresolve_provider_config
( model: &'a str, custom_llm_provider: Option<&'a str>, )
litellm-rust/crates/inference-ocr/src/provider_config.rs:244
↓ 10 callersFunctionresolve_token_count_settings
( configured: Mapping[str, object], default: Mapping[str, object], deployment_region: Callable[[], str] )
litellm/proxy/common_utils/token_count_settings.py:200
↓ 10 callersFunctionresourceLiteLLMTeamRead
(d *schema.ResourceData, m interface{})
terraform/provider/litellm/resource_team.go:208
↓ 10 callersMethodresponse_api_handler
Handles responses API requests. When _is_async=True, returns a coroutine instead of making the call directly. Keeps the pre-
litellm/llms/custom_httpx/llm_http_handler.py:3000
↓ 10 callersFunctionresponse_id
(response: httpx.Response)
tests/integration/caching/response_cache_case.py:370
↓ 10 callersMethodresponse_scans
(self, secret: str)
tests/integration/observability/test_guardrail_effects.py:2404
↓ 10 callersFunctionresponse_tool_call_names
Tool names invoked in a completion response, in call order, for any response surface get_tool_calls_from_response understands (chat completions, R
litellm/proxy/db/spend_log_tool_index.py:48
↓ 10 callersFunctionresult
(index: int, user_id: str)
litellm/proxy/management_helpers/bulk_user_deletion.py:723
↓ 10 callersMethodresume
(self, ctx: SessionContext, native_session_id: str)
tests/unit/harness/core_fakes.py:194
↓ 10 callersMethodresume
(&mut self)
litellm-rust/crates/coroutine/src/coroutine.rs:56
↓ 10 callersFunctionretryCredentialRead
retryCredentialRead attempts to read a credential with exponential backoff. If the read path clears the ID (e.g., transient 404 right after create), w
terraform/provider/litellm/resource_credential_crud.go:25
↓ 10 callersMethodreview
( self, submission_id: str, reviewer: str, approve: bool, reason: str
litellm/models/skill_submission.py:102
↓ 10 callersFunctionrf
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 10 callersMethodroute
Route a requested model to the corresponding llm deployments based on the regex pattern loop through all the patterns and find the m
litellm/litellm_core_utils/pattern_match_deployments.py:156
↓ 10 callersMethodrouter_cooldowns
(self)
litellm/redis/stores/connection.py:27
↓ 10 callersFunctionrun
( container: &ContainerAsync<I>, command: Vec<String>, )
litellm-rust/crates/redis-testing/src/exec.rs:8
↓ 10 callersFunctionrun_child_interpreter
Run `source` in a fresh interpreter that imports the same `litellm` as this process. `-I` keeps the working directory off sys.path so a source ch
tests/test_litellm_rust/support/child_interpreter.py:20
↓ 10 callersFunctionsafe_get
Fetch a user-supplied URL with SSRF protection on every redirect hop. Validates the initial URL and each redirect target before making the
litellm/litellm_core_utils/url_utils.py:460
↓ 10 callersFunctionsave_settings
( repository: ConfigRepository, settings: ROISettings, encrypted_token: str, encrypted_estimat
litellm/proxy/roi_calculator/settings.py:188
↓ 10 callersFunctionscript_printing
(output: &str)
litellm-rust/crates/testkit/tests/support/mod.rs:46
↓ 10 callersFunctionscripted_routes
( create: JsonResponse, read: JsonResponse, *, provider_id: str = PROVIDER_ID, **extra: JsonResponse )
tests/integration/spend/_background_responses.py:87
↓ 10 callersFunctionsecret_names
Every environment name the Vertex params and the token acquisition read, for hosts that resolve secrets up front.
litellm-rust/crates/auth-gcp/src/constants.rs:15
↓ 10 callersFunctionseeded_row
(table: str, entity_column: str, entity: str | None, api_key: str, date: str)
tests/integration/_support/daily_activity.py:235
← previousnext →7,201–7,300 of 202,479, ranked by callers