MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 7 callersMethodmap_team_model
( self, team_model_name: str | None, team_id: str, custom_llm_provider: str | None = None )
litellm/router_deprecated.py:542
↓ 7 callersMethodmarkDirty
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 7 callersFunctionmarker_of
(request: Request)
tests/integration/providers/_tool_search_support.py:84
↓ 7 callersMethodmarks
()
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 7 callersFunctionmay_name_litellm_mcp_gateway
(tools: Iterable[object] | None)
litellm/responses/mcp/tool_precheck.py:46
↓ 7 callersFunctionmember_ids
(candidate: Gateway, team: str)
tests/integration/management/test_scim_group_pathless_patch.py:63
↓ 7 callersMethodmemory
(self)
litellm-rust/crates/redis-store/src/auth_cache/types.rs:25
↓ 7 callersFunctionmerge_bedrock_aws_request_params
Merge deployment and request parameters without allowing auth escalation. Deployment configuration is authoritative for AWS authentication. When
litellm/llms/bedrock/common_utils.py:138
↓ 7 callersFunctionmerge_consecutive_system_messages
( messages: list[AllMessageValues], # mutable-ok: message pipelines type messages as mutable lists )
litellm/litellm_core_utils/prompt_templates/common_utils.py:1540
↓ 7 callersMethodmerge_tags
Helper function to merge two lists of tags, ensuring no duplicates. Args: request_tags (Optional[list]): List of tags fr
litellm/proxy/litellm_pre_call_utils.py:2292
↓ 7 callersFunctionmessages
(value: serde_json::Value)
litellm-rust/crates/core-utils/src/prompt_templates/factory.rs:134
↓ 7 callersFunctionmetadataPairsToObject
( pairs: readonly (Partial<MetadataPair> | undefined)[] | undefined, )
ui/litellm-dashboard/src/components/common_components/MetadataKeyValueFields.tsx:75
↓ 7 callersFunctionmetadata_bucket
``kwargs.setdefault(name, {})``: the request's own metadata dict, created when absent.
litellm/router_rust/invoke/shaping.py:98
↓ 7 callersMethodmget
(self, keys: Sequence[str])
tests/unit/caching/test_redis_batch.py:55
↓ 7 callersFunctionmigrator
(migrations: Vec<Migration>)
litellm-rust/crates/storage-clickhouse/tests/migrations.rs:76
↓ 7 callersFunctionmint_ephemeral_dcr_client
Mint a throwaway OAuth client via the upstream's RFC 7591 registration endpoint for a client-forwarded-token server whose authorize arrived with n
litellm/proxy/mcp_server/discoverable_endpoints.py:2073
↓ 7 callersFunctionmodelHubCall
(accessToken: string)
ui/litellm-dashboard/src/components/networking.tsx:1782
↓ 7 callersFunctionmodelPatchUpdateCall
( accessToken: string, formValues: Record<string, any>, // Assuming formValues is an object modelId: str
ui/litellm-dashboard/src/components/networking.tsx:2586
↓ 7 callersMethodmodel_dump
(self)
tests/unit/llms/xinference/image_generation/test_xinference_image_generation.py:22
↓ 7 callersFunctionmodel_list_reply
()
tests/integration/routing/test_payment_required_mapping.py:44
↓ 7 callersMethodmodel_supports_effort_param
( cls, model: str, custom_llm_provider: str, )
litellm/llms/anthropic/chat/transformation.py:755
↓ 7 callersFunctionmonitor_spend_logs_queue
Background task that monitors the spend_log_transactions queue size and triggers processing when the threshold is reached. Args:
litellm/proxy/utils.py:9270
↓ 7 callersMethodmove
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 7 callersFunctionn
()
litellm/proxy/_experimental/out/_next/static/chunks/0dvgjd47fte1d.js:1
↓ 7 callersFunctionname
(ask: Ask)
litellm-rust/crates/coroutine/tests/coroutine.rs:35
↓ 7 callersMethodnamed
`_core_deployments_named(name)`: the deployments the model-name index holds for `name` (whatever their own name), or `None` when it does not index the
litellm-rust/crates/router-core/src/registry.rs:749
↓ 7 callersMethodnames
( self, model_name: str | None = None, *, kind: Literal["groups", "ids"] = "gr
litellm/router_api.py:641
↓ 7 callersMethodnative
The request's native batches, built on first use so requests that touch no native store pay nothing. They keep a copy of the context they are
litellm/caching/redis_batch.py:575
↓ 7 callersFunctionnative_redacted_thinking_data
Claude's own `data` string for a redacted_thinking block. Bedrock Converse carries it base64-wrapped as `redactedContent`, and LiteLLM passed tha
litellm/litellm_core_utils/prompt_templates/reasoning_utils.py:227
↓ 7 callersMethodnative_session_id
(self)
tests/unit/harness/core_fakes.py:191
↓ 7 callersFunctionnative_stream_frames
( message: Mapping[str, JsonValue], *, delta_extra: Mapping[str, JsonValue] | None = None )
tests/integration/_support/safeguards_rig.py:649
↓ 7 callersFunctionnewGuardrailTestData
(t *testing.T, raw map[string]interface{})
terraform/provider/litellm/resource_guardrail_test.go:13
↓ 7 callersFunctionnewTestResourceData
newTestResourceData creates a *schema.ResourceData with the credential schema, sets the ID and populates the required fields.
terraform/provider/litellm/resource_credential_crud_test.go:21
↓ 7 callersMethodnext_event
(self)
tests/unit/test_router/test_router_provider_endpoints.py:80
↓ 7 callersFunctionnormalize
Lowercase, NFKC, strip zero-width, leetspeak, collapse spaces.
litellm/proxy/guardrails/guardrail_hooks/litellm_content_filter/competitor_intent/base.py:23
↓ 7 callersFunctionnormalizeArray
(s)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 7 callersFunctionnormalize_bedrock_opus_output_config_effort
Normalize Anthropic ``output_config.effort`` values for Bedrock Opus ids. Bedrock's Claude Opus request validator can accept a narrower effo
litellm/llms/bedrock/common_utils.py:1905
↓ 7 callersFunctionnormalize_email
Normalize email address to lowercase for consistent storage and comparison. Email addresses should be treated as case-insensitive for SSO pu
litellm/proxy/management_endpoints/ui_sso.py:881
↓ 7 callersMethodnormalize_mai_image_usage
Map Azure MAI usage fields to OpenAI ImageUsage schema.
litellm/llms/azure_ai/image_generation/mai_transformation.py:147
↓ 7 callersFunctionnormalize_request_params
Down-convert forwarded request ``params`` from the served version to 0.3. Upstream agents in this proxy pivot on 0.3 wire format, so when LiteLLM
litellm/proxy/a2a/version_convert.py:88
↓ 7 callersMethodnormalize_responses_api_request_dict
Apply provider-agnostic fixes to an outbound Responses API request dict.
litellm/llms/base_llm/responses/transformation.py:417
↓ 7 callersFunctionnormalize_route_for_root_path
Strip SERVER_ROOT_PATH prefix. Returns de-prefixed route, or None if route is not under root path.
litellm/proxy/utils.py:9967
↓ 7 callersFunctionnormalize_strategy
A strategy as the string Python compares against (``_normalize_strategy``).
litellm/router_rust/steps/selection.py:60
↓ 7 callersFunctionnot_yet_routable
(response: object)
tests/integration/providers/_coralbricks.py:339
↓ 7 callersFunctionnumber
(value: JsonValue)
tests/integration/providers/_coralbricks.py:52
↓ 7 callersFunctiono
litellm/proxy/_experimental/out/_next/static/chunks/11et800l-khs-.js:49
↓ 7 callersFunctiono
()
litellm/proxy/_experimental/out/_next/static/chunks/218jq7ygtsrfo.js:1
↓ 7 callersFunctiono
({tree:e})
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 7 callersFunctionoauth_protected_resource_path
The server's RFC 9728 Protected Resource Metadata path, the shared anchor of both challenges. ``root_path`` is the prefix the request was routed
litellm/proxy/mcp_server/outbound_credentials/adapter.py:334
↓ 7 callersFunctionobserve
( namespace: &Py<PyDict>, py: Python<'_>, expression: &str, )
litellm-rust/crates/host-python/src/gil.rs:120
↓ 7 callersMethodobserve
The routing observers: the Slack alerting logger ``alerting_config`` built, if any. ``request_scoped=True`` takes the router's callbacks off t
litellm/router_rust/core.py:2010
↓ 7 callersFunctionobserveResponseHeaders
(onHeaders: ResponseHeadersListener)
ui/litellm-dashboard/src/lib/http/observeResponseHeaders.ts:4
↓ 7 callersFunctionobserved
(gateway: Gateway)
tests/integration/routing/test_fallback_hop_foreign_encrypted_reasoning.py:352
↓ 7 callersFunctionocr
(request: pytest.FixtureRequest)
tests/unit/rust_bridge/ocr/test_secrets.py:100
↓ 7 callersFunctionocr_request_with_document
( model: &str, base: &str, document: Value, options: Value, )
litellm-rust/crates/inference-ocr/tests/ocr/main.rs:103
↓ 7 callersFunctionoe
(e,t,n,r)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 7 callersFunctionone_served_genai_span
(trace: JaegerTrace, genai_span: str)
tests/e2e/otel_client.py:151
↓ 7 callersFunctionopen_continuation
( operation: InteractionOperation, context: OperationContext, *, now: int )
litellm/proxy/mcp_server/interactions.py:148
↓ 7 callersFunctionopen_gateway_dcr_client
(client_id: str)
litellm/proxy/mcp_server/gateway_dcr_flow.py:511
↓ 7 callersFunctionopen_session_refresh_bearer
Open a session refresh token presented on a ``refresh_token`` grant. The token-endpoint mirror of :func:`resolve_session_bearer`: strips an optio
litellm/proxy/mcp_server/outbound_credentials/session_credentials.py:274
↓ 7 callersFunctionopen_session_refresh_token
Validate a session REFRESH ``candidate`` and recover the principal. Total over hostile input exactly like :func:`open_session_token`. The ``k
litellm/proxy/mcp_server/outbound_credentials/session_token.py:393
↓ 7 callersFunctionopenai_chat_refusal_text
Refusal text carried by an OpenAI Chat Completions message or streaming delta, read from ``refusal`` or from the ``provider_specific_fields``
litellm/llms/anthropic/messages/native/utils.py:153
↓ 7 callersFunctionopenapi_peer
OpenAPI-described HTTP service plus the spec file the proxy turns into MCP tools.
tests/integration/_support/mcp.py:402
↓ 7 callersMethodopenid_from_response
( response: dict | None, team_ids: list[str], user_role: LitellmUserRoles | None,
litellm/proxy/management_endpoints/ui_sso.py:4802
↓ 7 callersFunctionoutcome_wire_value
The client-visible form of one outcome, for the tools/list result ``_meta`` and the REST response: category, status code and controlled diagnostic
litellm/proxy/mcp_server/faults/list_outcomes.py:325
↓ 7 callersMethodoutput
(self)
litellm/utils.py:6615
↓ 7 callersFunctionoverlong_install_paths
(wheel, prefix_len=WORST_CASE_PREFIX)
tests/windows_tests/check_windows_wheel_install.py:39
↓ 7 callersFunctionp
litellm/proxy/_experimental/out/_next/static/chunks/11et800l-khs-.js:1
↓ 7 callersFunctionp
(s,r)
litellm/proxy/_experimental/out/_next/static/chunks/1f2vcab0yti03.js:53
↓ 7 callersFunctionp
()
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 7 callersFunctionparse
(variables: &[(&str, &str)])
litellm-rust/crates/gateway/src/settings.rs:83
↓ 7 callersFunctionparse_all
(name: str, state: PiStreamState | None = None)
tests/unit/llms/pi/harness/test_transformation.py:75
↓ 7 callersFunctionparse_budget_reset_time
Parse a `budget_reset_time` config value (e.g. "12:00") into a `time`. Falls back to midnight when unset; raises a clear error on a malformed val
litellm/proxy/common_utils/timezone_utils.py:24
↓ 7 callersFunctionparse_database_token_expiration
Return when ``token`` expires as a naive UTC datetime, or None when unreadable. Callers fall back to a fixed refresh interval on None, so an unpa
litellm/proxy/db/token_auth.py:198
↓ 7 callersFunctionparse_iam_endpoint_from_url
Parse an :class:`IAMEndpoint` back out of a Postgres URL. Used so a reader URL can drive its own token refresh without requiring callers to s
litellm/proxy/db/token_auth.py:118
↓ 7 callersFunctionparse_jsonl_with_embedded_newlines
Parse JSONL content that may contain JSON objects with embedded newlines in string values. Unlike splitlines(), this function properly handl
litellm/litellm_core_utils/batch_utils.py:45
↓ 7 callersFunctionparse_model_fields
(schema: str)
scripts/check_schema_compatibility.py:19
↓ 7 callersFunctionparse_reload_schedule
(row: "LiteLLM_Config")
litellm/proxy/common_utils/periodic_reload_schedule.py:112
↓ 7 callersFunctionparse_request
Read a request; `model` replaces the request's own when given.
litellm-rust/crates/python-bridge/src/router/config.rs:224
↓ 7 callersMethodparse_search_tools
Parse and validate search tools from config. Loads environment variables and casts to SearchToolTypedDict. Args:
litellm/proxy/proxy_server.py:6669
↓ 7 callersFunctionparse_systemone_response
(payload: object, request: DecisionsIRRequest)
litellm/llms/base_llm/decisions/transformation.py:438
↓ 7 callersFunctionparse_transcription_session_update
( payload: str, error: ProtocolErrorType = RealtimeTranscriptionProtocolError )
litellm/llms/base_llm/realtime/transcription_protocol.py:119
↓ 7 callersFunctionparse_upstream_reported_usage
Read the reported totals off an upstream pass-through response. Returns ``None`` when neither header is present, which is the normal case for
litellm/proxy/pass_through_endpoints/upstream_usage_headers.py:104
↓ 7 callersFunctionparse_variations_response
Parse the LLM response for competitor variations into a name -> variations map.
litellm/proxy/management_endpoints/policy_endpoints/endpoints.py:967
↓ 7 callersMethodparse_verdict
No verdict allows; a failed or unreadable reply raises so unreachable_fallback decides.
litellm/proxy/guardrails/guardrail_hooks/akto/akto.py:621
↓ 7 callersMethodpartition_index_name
The child index name for one partition, built the way Postgres names the children of a partitioned index, and kept within the 63 byte identifi
litellm-proxy-extras/litellm_proxy_extras/request_log_indexes.py:47
↓ 7 callersFunctionpatch_capacity_snapshot
Stub :func:`cassette_cache_capacity_snapshot` so we don't try to open a real Redis connection. The fixture returns a setter that each test use
tests/unit/test_vcr_conftest_common_banner.py:55
↓ 7 callersFunctionpatch_direct_endpoints
Replace litellm's directly bound endpoint functions with recording stand-ins on ``host``. The stand-in keeps the function's ``__name__`` (the cor
tests/unit/test_router/core/endpoints_helpers.py:182
↓ 7 callersFunctionpatch_group
Patch a group according to SCIM v2 protocol
litellm/proxy/management_endpoints/scim/scim_v2.py:3446
↓ 7 callersMethodpaused
(self)
tests/integration/_support/redis_process.py:126
↓ 7 callersMethodpeek
(&self)
litellm-rust/crates/python-compat/src/json.rs:84
↓ 7 callersFunctionpem_of
(key: ec.EllipticCurvePrivateKey)
tests/unit/llms/base_llm/auth/test_internal_issuer.py:29
↓ 7 callersFunctionpieces
`text` cut into the deltas a provider streams it as, `size` characters each.
tests/e2e/perf/setup/mocks/reply_common.py:159
↓ 7 callersFunctionpin_config
( directory: Path, model_list: Sequence[JsonValue], *, litellm_settings: Mapping[str, JsonValu
tests/integration/routing/_prompt_cache_pin_support.py:250
↓ 7 callersFunctionpipeline
( authority: &'static str, permissions: Permissions, restrictions: Permissions, failure: Optio
litellm-rust/crates/gateway-auth/tests/auth.rs:209
↓ 7 callersFunctionpipeline_managed_guardrail_names
( data: Mapping[str, object], mode: Literal["pre_call", "post_call"] )
litellm/proxy/utils.py:798
↓ 7 callersFunctionplacement
(api: GitHubApi, repo: string, mergeCommit: string)
scripts/comment-fixed-issue.ts:252
↓ 7 callersFunctionpod_reload_is_due
Whether this pod should reload now. A revision it has not applied means a manual reload it has not served. A pod starts at revision 0, so it
litellm/proxy/common_utils/periodic_reload_schedule.py:138
↓ 7 callersMethodpoll
(e,t,r)
litellm/proxy/_experimental/out/_next/static/chunks/2hk8n3azev--d.js:14
↓ 7 callersFunctionpool
()
litellm-rust/crates/http/src/pool.rs:145
← previousnext →11,101–11,200 of 202,479, ranked by callers