MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 12 callersFunctionread_request
(base: &str, options: Value)
litellm-rust/crates/inference-ocr/tests/ocr/azure_document_intelligence.rs:17
↓ 12 callersMethodrecord_partial_usage_for_failure
Stash what an interrupted stream already consumed so the failure log bills it instead of zero.
litellm/litellm_core_utils/litellm_logging.py:2700
↓ 12 callersFunctionrecorded
(samples: Sequence[Sample], **labels: str)
tests/integration/observability/test_prometheus_provider_auth_failure.py:462
↓ 12 callersMethodrecords
(self)
tests/unit/integrations/pointfive/test_logger.py:40
↓ 12 callersFunctionredact_internal_details_from_client_message
Public API: redact_secrets() plus filesystem paths, internal hostnames, and an embedded traceback, for a string about to leave the process in an H
litellm/_logging.py:137
↓ 12 callersFunctionredis_key_for
(cassette_path: str)
tests/_vcr_redis_persister.py:95
↓ 12 callersMethodregister_tool
Register a new tool in the registry
litellm/proxy/mcp_server/tool_registry.py:55
↓ 12 callersFunctionregistered
The ``litellm.model_cost`` entry registered under a fresh deployment id.
tests/unit/test_router/core/test_gaps_l8400_pricing.py:21
↓ 12 callersFunctionrelease_langfuse_tracing
Let go of one logger's hold on its channel; a channel nobody holds is retired ``grace_seconds`` later. The grace covers a callback that fetched i
litellm/integrations/langfuse/langfuse_sdk.py:940
↓ 12 callersFunctionrender_refusal
(refusal: UnsafeMasterKeyRefused)
litellm/proxy/auth/master_key_boot_check.py:149
↓ 12 callersFunctionrequest_body
(kwargs: dict[str, object])
tests/test_litellm_rust/support/callback_recorder.py:149
↓ 12 callersFunctionrequest_from_call
The call's arguments as one keyword request. Raises ``TypeError`` like the endpoint would.
litellm/router_attempt.py:73
↓ 12 callersFunctionresolve_bucket_includes
( *, config: Mapping[str, object], object_key: str, fetch: BucketObjectFetcher, )
litellm/proxy/common_utils/load_config_utils.py:139
↓ 12 callersMethodresolve_credentials
(self, auth_params: AwsAuthParams, aws_region_name: str | None)
litellm/llms/bedrock/base_aws_llm.py:463
↓ 12 callersFunctionresolve_image_model_info
The price table an image cost calculator consults for ``model``. ``shared_backend_model_info`` keeps deployment prices off the shared ``{provider
litellm/litellm_core_utils/llm_cost_calc/utils.py:998
↓ 12 callersFunctionresolve_resource_owner_id
Return the identity to stamp on (and match against) a managed resource's ``created_by``. A key with neither a user_id nor a team_id would oth
litellm/llms/base_llm/managed_resources/isolation.py:25
↓ 12 callersFunctionresolve_team_admin_editable_fields
( general_settings: Mapping[str, object], supported: frozenset[str], )
litellm/proxy/management_endpoints/team_admin_field_permissions.py:166
↓ 12 callersMethodresolve_toolset_tool_permissions
Resolve a list of toolset IDs into a mcp_tool_permissions dict. Returns: {server_id: [tool_name, ...]} — the union of all tools acro
litellm/proxy/mcp_server/mcp_server_manager.py:3823
↓ 12 callersFunctionresolve_validated_local_image_path
Resolve ``candidate`` only when it is an existing supported image file.
litellm/proxy/common_utils/static_asset_utils.py:26
↓ 12 callersMethodresolver
(self)
tests/unit/proxy/auth/test_resolvers_grants.py:58
↓ 12 callersFunctionresourceLiteLLMModel
()
terraform/provider/litellm/resource_model.go:8
↓ 12 callersFunctionresponse_body
(status: str, *, background: bool = True, **fields: JsonValue)
tests/integration/spend/_background_responses.py:36
↓ 12 callersFunctionresponse_cache_phase
The ``cache.get llm_response`` / ``cache.set llm_response`` span a response-cache read or write runs inside, so its datastore spans nest under it
litellm/caching/caching.py:81
↓ 12 callersFunctionresponse_payload
()
tests/unit/llms/bigmodel/test_bigmodel_provider.py:66
↓ 12 callersMethodrows
(self)
tests/integration/observability/_ternary_support.py:64
↓ 12 callersFunctionrun
(rollout: Rollout, calls: Recorder, *, native_missing: bool = False, context: RouteContext = CONTEXT)
tests/unit/rust_bridge/test_runtime.py:65
↓ 12 callersMethodrun
( self, args: tuple[object, ...], kwargs: Mapping[str, object], *, pyt
litellm/rust_bridge/dispatch.py:51
↓ 12 callersFunctionsafe_deep_copy
Safe Deep Copy The LiteLLM request may contain objects that cannot be pickled/deep-copied (e.g., tracing spans, locks, clients). Th
litellm/litellm_core_utils/core_helpers.py:785
↓ 12 callersMethodsave_cassette
(cassette_path, cassette_dict, serializer)
tests/_vcr_redis_persister.py:187
↓ 12 callersMethodschedule_acompletion
( self, model: str, messages: object, priority: int | None = None, str
litellm/router_rust/invoke/endpoints.py:2008
↓ 12 callersMethodscores
Cosine similarity for embeddable entries of `texts`, with None for omitted entries.
litellm/proxy/common_utils/semantic_text_index.py:289
↓ 12 callersFunctionsearch_mcp_tools
( query: str, tools: Sequence[Tool], top_k: int, settings: MCPToolSearchSettings, ranker:
litellm/proxy/mcp_server/tool_search.py:316
↓ 12 callersFunctionsend
( proxy: Gateway, surface: str, outcome: str, call_id: str, *, key: str | None = None, stream: bool = Fals
tests/integration/_support/hook_counts.py:258
↓ 12 callersFunctionseries_cap_rig
( directory: Path, settings: Mapping[str, JsonValue], *, workers: int, warm_keys: int,
tests/integration/_support/prometheus_series.py:177
↓ 12 callersFunctionsetSearch
(search: string)
ui/litellm-dashboard/src/app/login/LoginPage.integration.test.tsx:68
↓ 12 callersMethodset_attribute
(self, key: str, value: str)
litellm/integrations/otel/mount.py:107
↓ 12 callersMethodset_llm_deployment_failure_metrics
Sets Failure metrics when an LLM API call fails - mark the deployment as partial outage - increment deployment failure respo
litellm/integrations/prometheus.py:3228
↓ 12 callersFunctionset_trace_id
(trace_id: str)
litellm/_logging.py:71
↓ 12 callersFunctionsettle_candidate
( candidate: Gateway, model: str, key: str, status: int, *, burst: int = GATEWAY_BURST
tests/integration/authorization/_hidden_alias_budget.py:173
↓ 12 callersMethodsettled_akto_calls
(self, response_id: str, marker: str, key: str)
tests/integration/observability/test_akto_logging_only.py:242
↓ 12 callersMethodsettled_cost
(self)
tests/unit/llms/sail/helpers.py:115
↓ 12 callersMethodsetup_session_update_rejection
( self, model: str, session: Mapping[str, object], *, setup_sent: bool )
litellm/llms/gemini/realtime/transformation.py:575
↓ 12 callersFunctionsi
(e,t,n)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 12 callersFunctionsingle_flight_bearer_token_provider
( credential: "TokenCredential | SupportsTokenInfo", scope: str )
litellm/secret_managers/get_azure_ad_token_provider.py:115
↓ 12 callersFunctionsource
Where a Rust route reads provider secrets from. Python's `get_secret_str` when the settings projection reports no configured secret manager, otherwise
litellm-rust/crates/python-bridge/src/secrets/mod.rs:30
↓ 12 callersFunctionstart_replicas
( stack: ExitStack, containers: Containers, database: Database, migrations: tuple[Migration, ...] = (), co
tests/e2e/migrations/checks.py:28
↓ 12 callersMethodstop
(self)
litellm/proxy/common_utils/config_sync_pubsub.py:282
↓ 12 callersFunctionstream
(name: str)
tests/integration/providers/test_databricks_agent_wire.py:235
↓ 12 callersFunctionstreaming
(call: MessagesCall, api_base: String)
litellm-rust/crates/inference-messages/tests/messages/stream.rs:128
↓ 12 callersMethodsubtract
(s,o)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 12 callersFunctionsuccesses
(core: RouterCore, deployment_id: str)
tests/unit/test_router/core/test_streaming_characterization_outcomes.py:84
↓ 12 callersFunctionsummarize
One row per (coroutine, location) over the hold threshold, worst total first.
tests/e2e/perf/setup/event_loop.py:163
↓ 12 callersFunctiont
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/0g3hfgh2z8the.js:1
↓ 12 callersFunctiontag_cache_key
Cache key one tag row is stored under; shared so its five reader/writer modules cannot drift.
litellm/proxy/common_utils/user_api_key_cache.py:521
↓ 12 callersFunctionteam_deployment
A team-owned deployment: ``model_info.team_id`` and ``team_public_model_name`` set.
tests/unit/test_router/core/registry_helpers.py:32
↓ 12 callersFunctionteam_model_aliases
(team_object: LiteLLM_TeamTable | None)
litellm/proxy/auth/team_grants.py:170
↓ 12 callersFunctiontext_of
(value: JsonValue)
tests/integration/providers/_gemini_long_decode_support.py:285
↓ 12 callersMethodthaw
(&self)
litellm-rust/crates/redis-testing/src/cluster.rs:157
↓ 12 callersFunctionto_ns
Coerce a datetime / epoch value to integer nanoseconds.
litellm/integrations/otel/model/utils.py:81
↓ 12 callersFunctiontool_calls_with_server_tool_calls
( message: object, )
litellm/llms/anthropic/common_utils.py:2326
↓ 12 callersMethodtransaction
()
litellm/proxy/db/baseline_accounting.py:430
↓ 12 callersMethodtransform_delete_file_request
( self, file_id: str, optional_params: Mapping[str, object], litellm_params: M
litellm/llms/xai/files/transformation.py:186
↓ 12 callersMethodtransform_google_generate_content_to_openai_model_response
( self, completion_response: GenerateContentResponseBody | dict[str, object], # mutable-ok: m
litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py:3355
↓ 12 callersMethodtransform_interactions_usage_object
(usage_object: Mapping[str, object])
litellm/litellm_core_utils/llm_cost_calc/usage_object_transformation.py:133
↓ 12 callersMethodtransform_realtime_response
Transform Bedrock Nova Sonic response to OpenAI realtime format. Args: message: Bedrock format message (JSON string)
litellm/llms/bedrock/realtime/transformation.py:1609
↓ 12 callersMethodtransform_request
( self, model: str, messages: list[AllMessageValues], optional_params: dict,
litellm/llms/ollama/chat/transformation.py:233
↓ 12 callersMethodtransform_response
( self, model: str, raw_response: Response, model_response: ModelResponse,
litellm/llms/predibase/chat/transformation.py:145
↓ 12 callersMethodtransform_response
Transform Responses API response to chat completion response
litellm/completion_extras/litellm_responses_transformation/transformation.py:1622
↓ 12 callersMethodtransform_user_api_key_dict_to_metadata
Transform user_api_key_dict to a metadata dict with prefixed keys. Converts keys like 'user_id' to 'user_api_key_user_id' to clearly
litellm/llms/base_llm/guardrail_translation/base_translation.py:89
↓ 12 callersMethodtransform_vertex_ai_batch_response_to_openai_batch_response
( cls, response: VertexBatchPredictionResponse )
litellm/llms/vertex_ai/batches/transformation.py:241
↓ 12 callersMethodtransform_video_content_response
( self, raw_response: httpx.Response, logging_obj: LiteLLMLoggingObj, client:
tests/integration/_support/custom_video_provider.py:116
↓ 12 callersMethodtranslate_developer_role_to_system_role
( self, messages: Sequence[AllMessageValues], *, model: str, custom_ll
litellm/llms/openai/openai.py:222
↓ 12 callersFunctiontype_to_response_format_param
Re-implementation of openai's 'type_to_response_format_param' function Used for converting pydantic object to api schema.
litellm/llms/base_llm/base_utils.py:485
↓ 12 callersFunctionunconfigure_codex_settings
( settings_path: Path, *, commit: Callable[[str, str], None] = commit_staged_json )
litellm/proxy/client/cli/commands/codex_settings.py:275
↓ 12 callersFunctionuntil
(description: str, condition: Callable[[], bool], seconds: float = 150)
tests/e2e/migrations/containers.py:32
↓ 12 callersMethodupdate
(self, *, where: Mapping[str, object], data: Mapping[str, object])
litellm/proxy/auth/auth_checks.py:270
↓ 12 callersMethodupdate
(self, where: Mapping[str, object], data: Mapping[str, object])
litellm/proxy/search_endpoints/search_tool_registry.py:44
↓ 12 callersFunctionupdate_group
Update a group according to SCIM v2 protocol
litellm/proxy/management_endpoints/scim/scim_v2.py:3032
↓ 12 callersMethodupdate_in_memory_litellm_params
(self, litellm_params)
tests/unit/proxy/guardrails/guardrail_hooks/test_presidio.py:837
↓ 12 callersMethodupdate_values
( self, alerting: list | None = None, alerting_threshold: float | None = None,
litellm/proxy/utils.py:2085
↓ 12 callersFunctionuseMCPServers
(teamId?: string | null, userId?: string | null)
ui/litellm-dashboard/src/app/(dashboard)/hooks/mcpServers/useMCPServers.ts:9
↓ 12 callersFunctionuse_custom_pricing_for_model
Check if the model uses custom pricing Returns True if any custom pricing field is present in `litellm_params`, or if any custom pricing
litellm/litellm_core_utils/litellm_logging.py:6857
↓ 12 callersFunctionusesLlmClassifier
(classifierType: ClassifierType)
ui/litellm-dashboard/src/components/add_model/classifier_types.ts:17
↓ 12 callersFunctionvalidate_chat_completion_tool_choice
Confirm the tool choice is passed in the OpenAI format. Prevents user errors like: https://github.com/BerriAI/litellm/issues/7483
litellm/utils.py:9746
↓ 12 callersMethodvalidate_environment
( self, headers: dict, model: str, messages: list[AllMessageValues], o
litellm/llms/fireworks_ai/chat/transformation.py:224
↓ 12 callersFunctionvalidate_managed_id_requirement
Enforce proxy-level managed resources on every route that accepts a provider-issued id when ``litellm.require_managed_files`` is enabled, and
litellm/proxy/openai_files_endpoints/common_utils.py:1253
↓ 12 callersFunctionvalidate_password_not_breached
Raise ``ProxyException`` (400) if ``password`` appears in a known data breach. Fails open: an unreachable or misbehaving HIBP allows the password
litellm/proxy/auth/password_policy.py:173
↓ 12 callersFunctionvalidate_password_policy
Raise ``ProxyException`` (400) if ``password`` fails the configured policy.
litellm/proxy/auth/password_policy.py:97
↓ 12 callersFunctionw
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/3yrgf44a9g52e.js:1
↓ 12 callersMethodwait_for_drain
Poll the in-flight request counter until it reaches the drain target or ``timeout`` seconds elapse. Args: timeou
litellm/proxy/shutdown/graceful_shutdown_manager.py:125
↓ 12 callersFunctionwatched_spans
(collector: Wire)
tests/integration/observability/_langfuse_otel_metadata_support.py:299
↓ 12 callersFunctionwe
(t,e)
litellm/proxy/_experimental/out/_next/static/chunks/0cz1d0mv5g_q7.js:1
↓ 12 callersFunctionwhere_sql
`WHERE` body and its bind parameters, numbered from `first_index`. Returns `("", ())` when there is nothing to filter on. Every caller-supplied v
litellm/proxy/list_api/list_framework.py:282
↓ 12 callersMethodwithOptions
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2hk8n3azev--d.js:14
↓ 12 callersFunctionwrap_sse_stream_with_keepalive_pings
Fill idle gaps in an SSE stream, including the one before its first chunk. ``ping_chunk`` is what gets written into those gaps. It defaults to An
litellm/proxy/common_utils/sse_keepalive.py:45
↓ 12 callersMethodwrite
(self, data: bytes | bytearray | memoryview)
tests/unit/proxy/guardrails/guardrail_hooks/test_presidio.py:5280
↓ 12 callersFunctionwrite_field
Writes one string field back into a dict or an object. Pairs with read_field.
litellm/proxy/guardrails/guardrail_hooks/llm_shield_proxy/payload.py:114
↓ 12 callersFunctiony
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
↓ 12 callersFunctiony
(t)
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 11 callersFunctionC
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
← previousnext →5,701–5,800 of 202,479, ranked by callers