MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 9 callersMethodresolve
(self, source_root: Path)
tests/rust-python-harness/shared/tracing/pytest_usage.py:62
↓ 9 callersFunctionresolveComplexityDefaultModel
(value: ActiveTierSet, pinned?: string)
ui/litellm-dashboard/src/components/add_model/tier_rows.ts:93
↓ 9 callersFunctionresolve_embedding_router
Return ``llm_router`` iff it serves ``embedding_model`` as a deployment.
litellm/caching/_embedding_router.py:25
↓ 9 callersFunctionresolve_llm_passthrough_timeout
Resolve upstream httpx timeout for SDK native passthrough (e.g. Bedrock /converse, Anthropic /v1/messages). Non-streaming precedence: kw
litellm/passthrough/timeout_utils.py:45
↓ 9 callersFunctionresolve_model_and_provider_for_metadata
(model: str, custom_llm_provider: str | None = None)
litellm/litellm_core_utils/get_llm_provider_logic.py:167
↓ 9 callersFunctionresolve_request
Resolve `name` for `caller`. `None` when nothing serves it (`no_healthy_deployments`).
litellm-rust/crates/router-core/src/resolve.rs:297
↓ 9 callersFunctionresolve_valid_user_oauth_token
Return an OAuth2 credential whose access_token is good for the next request. Returns the credential unchanged while its token is valid for at lea
litellm/proxy/mcp_server/db.py:2180
↓ 9 callersFunctionresourceLiteLLMJWTKeyMappingCreate
(d *schema.ResourceData, m interface{})
terraform/provider/litellm/resource_jwt_key_mapping_crud.go:15
↓ 9 callersMethodrespond
(request: httpx.Request)
tests/unit/llms/minimax/videos/test_minimax_video_transformation.py:1155
↓ 9 callersFunctionresponse_coro
()
tests/unit/passthrough/test_streaming_interrupt_spend_tracking.py:73
↓ 9 callersFunctionresponse_header
Typed read of an SDK response header: httpx.Headers.get returns Any and httpx itself is a banned import in suite code, so tests read headers throu
tests/e2e/llm_translation/sdk_clients.py:32
↓ 9 callersFunctionresponses_api
Follows the OpenAI Responses API spec: https://platform.openai.com/docs/api-reference/responses Supports background mode with polling_via_ca
litellm/proxy/response_api_endpoints/endpoints.py:316
↓ 9 callersFunctionresponses_websocket_endpoint
Responses API WebSocket mode endpoint. Keeps a persistent WebSocket connection for response.create events, enabling lower-latency agenti
litellm/proxy/response_api_endpoints/endpoints.py:2721
↓ 9 callersFunctionrestore_claude_settings
Restore settings_path from the backup at backup_path, then delete the backup. Returns the restored record, or None if there was nothing to restor
litellm/proxy/client/cli/commands/up.py:91
↓ 9 callersMethodrestore_encrypted_content_item_ids_in_input
Decode litellm-encoded item IDs in request input back to original IDs. Called before forwarding the request to the upstream provider so the
litellm/responses/utils.py:634
↓ 9 callersFunctionrestricted_key
(monkeypatch: pytest.MonkeyPatch)
tests/unit/proxy/pass_through_endpoints/test_codex_passthrough_endpoints.py:604
↓ 9 callersFunctionrewrite_path_ids
Walk URL path segments and resolve any passthrough managed IDs to raw provider IDs. Returns *path* unchanged when no managed IDs are found.
litellm/proxy/pass_through_endpoints/managed_id_rewriter.py:1227
↓ 9 callersFunctionrfc7638_thumbprint
RFC 7638: SHA-256 over the lexicographically member-ordered, whitespace-free JSON rendering of the thumbprint members, base64url-encoded without p
litellm/llms/base_llm/auth/jwt_signing.py:71
↓ 9 callersMethodroute_type
(self)
tests/unit/proxy/video_endpoints/test_endpoints.py:138
↓ 9 callersFunctionrouter
(config: Value, dynamic: Value)
litellm-rust/crates/python-bridge/src/router/host.rs:1151
↓ 9 callersFunctionrouter
( inference: Arc<Gateway>, config: &Config, ui: Option<Router>, mcp: Option<Router>, )
litellm-rust/crates/gateway/src/lib.rs:139
↓ 9 callersMethodrows
(&self)
litellm-rust/crates/model-catalog/src/catalog.rs:433
↓ 9 callersFunctionrule
(sibling: unknown)
ui/litellm-dashboard/src/utils/ptuValidation.test.ts:78
↓ 9 callersMethodrun
Main loop: accept ``response.create`` events sequentially and handle each one before waiting for the next message.
litellm/responses/streaming_iterator.py:3701
↓ 9 callersFunctionrun_sync
Run coro on the harness loop thread and block for its result.
litellm/harness/sync.py:80
↓ 9 callersFunctionrx
(e,t,n,r)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 9 callersFunctions3_config
( path: Path, sink_url: str, extra: Mapping[str, JsonValue], settings: Mapping[str, JsonValue] | None = No
tests/integration/observability/_s3_v2_support.py:84
↓ 9 callersFunctionsafe_filename
Extract a safe filename from a user-supplied path. Strips all directory components (both Unix and Windows separators), returning only th
litellm/litellm_core_utils/path_utils.py:73
↓ 9 callersFunctionsample_report
``None`` when the consent sends nothing at all
litellm/telemetry/sample.py:87
↓ 9 callersFunctionsanitize_for_log
(value: object)
litellm/proxy/litellm_pre_call_utils.py:313
↓ 9 callersMethodsave
Save frames as optimized GIF for Slack. Args: output_path: Where to save the GIF num_colors: Number of color
tests/unit/skills/test_skills_data/slack-gif-creator/core/gif_builder.py:160
↓ 9 callersFunctionsave_config
Save CLI config to file
litellm/proxy/client/cli/commands/config.py:46
↓ 9 callersFunctionsave_grant
( repository: ConfigRepository, config: OAuthConfig, grant: TokenGrant, *, revision: int |
litellm/proxy/roi_calculator/oauth.py:269
↓ 9 callersFunctionseal_continuation
( result: BoundInputRequiredResult, operation: InteractionOperation, context: OperationContext,
litellm/proxy/mcp_server/interactions.py:168
↓ 9 callersFunctionsend
(group: str, **media: object)
tests/unit/decisions/test_main.py:4100
↓ 9 callersMethodsend
(e,t,s)
litellm/proxy/_experimental/out/_next/static/chunks/1f2vcab0yti03.js:28
↓ 9 callersMethodsend
( self, path: str, *, headers: BaseModel, json: BaseModel, par
tests/e2e/transport.py:53
↓ 9 callersFunctionsendChatCompletion
(request: APIRequestContext, opts: ChatOptions)
tests/e2e/ui/helpers/traffic.ts:52
↓ 9 callersMethodsend_request
(self, method: HttpMethods, endpoint: Endpoints, body=None, params=None)
litellm/integrations/deepeval/api.py:81
↓ 9 callersMethodsend_text
(self, text: str)
litellm/rust_bridge/responses/websocket.py:17
↓ 9 callersFunctionserialize_records
Serialize each record to one JSON line.
litellm/integrations/pointfive/payload.py:14
↓ 9 callersFunctionserve
( status_line: &'static str, )
litellm-rust/crates/http/src/pool.rs:163
↓ 9 callersMethodserve
Serves until `shutdown` resolves, then lets open requests and WebSockets finish for up to `drain` before closing whatever is left. Python shuts down a
litellm-rust/crates/gateway-forward/src/forward.rs:62
↓ 9 callersFunctionservice_events
(core: RouterCore)
tests/unit/test_router/core/test_parity_cooldown_service_model.py:576
↓ 9 callersMethodset
`model_cost[key] = {"litellm_provider": provider, ...}`
litellm-rust/crates/model-catalog/src/store/mod.rs:171
↓ 9 callersMethodset
`DualCache.async_set_cache`: memory, then a `SET` of `json.dumps(value)`. A Redis failure, or a value `json.dumps` refuses, only logs.
litellm-rust/crates/redis-store/src/dual_cache/write.rs:18
↓ 9 callersFunctionsetLocalStorageItem
(key: string, value: string)
ui/litellm-dashboard/src/utils/localStorageUtils.ts:17
↓ 9 callersMethodset_cache
( self, key: str, value: object, local_only: bool = False, **kwargs: o
litellm/proxy/utils.py:572
↓ 9 callersMethodset_cache
(self, key: str, value: str, *, ttl: int)
litellm/llms/databricks/common_utils.py:36
↓ 9 callersMethodset_cache
(&self, key: &str, value: C::Value, _: &ExactCacheContext)
litellm-rust/crates/cache-azure-blob/src/cache.rs:190
↓ 9 callersFunctionset_object_permission
Creates the LiteLLM_ObjectPermissionTable record for the key/team. Handles permissions for vector stores and mcp servers.
litellm/proxy/management_helpers/object_permission_utils.py:217
↓ 9 callersMethodset_preprocessing_duration_attribute
Set ``litellm.preprocessing.duration_ms`` (proxy-receive -> first provider handoff) on the proxy SERVER span. ``litellm_received_at``
litellm/integrations/opentelemetry.py:3670
↓ 9 callersFunctionset_session_id
(session_id: str)
litellm/_logging.py:67
↓ 9 callersMethodset_valid_azure_ad_token
Wrapper to set self.azure_auth_token to a valid Azure AD token, refreshing if necessary Without a service principal configured, the
litellm/integrations/azure_storage/azure_storage.py:307
↓ 9 callersMethodshadow_successes
(self)
tests/unit/test_router_silent_experiment.py:67
↓ 9 callersMethodshould_allow_request_on_db_unavailable
Returns True if the request should be allowed to proceed despite the DB connection error
litellm/proxy/db/exception_handler.py:118
↓ 9 callersMethodshould_use_cache
Returns true if we should use the cache for LLM API calls If cache is default_on then this is True If cache is default_off t
litellm/caching/caching.py:1021
↓ 9 callersMethodshutdown
Starts graceful shutdown and returns the server task, which ends once in-flight work drains or `DRAIN` passes.
litellm-rust/crates/gateway-forward/tests/support/mod.rs:78
↓ 9 callersFunctionsignature
(legacy: Callable[..., object])
litellm/rust_bridge/public_call.py:12
↓ 9 callersMethodsingle_span
(self, response_id: str, *, elsewhere: "Collector | None" = None)
tests/integration/observability/test_signoz_delivery.py:225
↓ 9 callersFunctionslow_upstream_async_client
(body: bytes = b"")
tests/unit/llms/bedrock/slow_upstream.py:47
↓ 9 callersMethodsnapshot
(self)
tests/unit/harness/core_fakes.py:99
↓ 9 callersFunctionspend_rows
(model: str, expected: int = 1)
tests/integration/providers/_gemini_long_decode_support.py:576
↓ 9 callersFunctionsplit_concatenated_json_objects
Split a string that contains one or more concatenated JSON objects into a list of parsed dicts. LLM providers (notably Bedrock Claude So
litellm/litellm_core_utils/prompt_templates/tool_call_json_repair.py:128
↓ 9 callersMethodsplit_text_by_words
Split text into chunks at word boundaries without breaking words. Always returns at least one chunk. Short text (≤ max_length) is
litellm/proxy/guardrails/guardrail_hooks/azure/base.py:97
↓ 9 callersFunctionsq
(e)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 9 callersFunctionsso_client_config
The configured SSO client the login used, or ``None`` when the deployment has none. Read from the process environment because that is where the l
litellm/proxy/mcp_server/outbound_credentials/sso_assertion_refresher.py:142
↓ 9 callersMethodstalled_within
(self, window_seconds: float)
litellm/proxy/db/db_lookup_gate.py:50
↓ 9 callersFunctionstamp_responses_usage_cost
The terminal response's ``usage.cost`` is the bill litellm writes to the spend row. It is priced on a copy shaped the way the success handler shap
litellm/responses/streaming_iterator.py:1898
↓ 9 callersMethodstart
( self, *, litellm_version: str, settings: TelemetrySettings, db: Call
litellm/proxy/telemetry/runtime.py:131
↓ 9 callersMethodstart
(self)
litellm/harness/sync.py:162
↓ 9 callersMethodstart
(self)
litellm/tracing/exporter.py:150
↓ 9 callersMethodstop
(self)
litellm/proxy/common_utils/auth_cache_invalidation_pubsub.py:262
↓ 9 callersFunctionstream_usage_relay
( endpoint: str, request_body: Mapping[str, object], general_settings: Mapping[str, object], s
litellm/proxy/pass_through_endpoints/injected_stream_usage.py:104
↓ 9 callersFunctionstreams
(request: Request)
tests/integration/_support/anthropic_thinking.py:146
↓ 9 callersMethodstrict_string
(&self)
litellm-rust/crates/python-bridge/src/coercion.rs:172
↓ 9 callersFunctionstripped_history
(marker: str, *reasoning: dict[str, JsonValue])
tests/integration/routing/test_fallback_hop_foreign_encrypted_reasoning.py:409
↓ 9 callersMethodsubscribe
( &self, )
litellm-rust/crates/redis-connect/src/refresh.rs:102
↓ 9 callersFunctionsummed_cost_breakdown
The breakdown of the charged rounds, or None when a charged round was priced without its own breakdown, since a partial one would disagree with th
litellm/responses/mcp/round_billing.py:119
↓ 9 callersFunctionsupports_anthropic_cache_control
( model: str, custom_llm_provider: str | None, model_info: Mapping[str, object] | None = None )
litellm/llms/anthropic/common_utils.py:333
↓ 9 callersFunctionsupports_prompt_caching
Check if the given model supports prompt caching and return a boolean value. Parameters: model (str): The model name to be checked.
litellm/utils.py:3603
↓ 9 callersFunctionsweep_all
S1 to S5 for one finished scenario; fails if any GET route returned no response. Redis goes first: it holds entries with a TTL, and the route wal
tests/integration/security/_sweeps.py:575
↓ 9 callersMethodsync_attachments_from_db
Sync policy attachments from the database to in-memory registry. Config-loaded attachments are preserved. Args:
litellm/proxy/policy_engine/attachment_registry.py:509
↓ 9 callersMethodsync_get_cache
( self, model: str, original_function: Callable[..., object], logging_obj: Lit
litellm/caching/caching_handler.py:514
↓ 9 callersMethodsync_user_role_and_teams
Sync user role and team memberships with JWT claims The goal of this method is to ensure: 1. The user role on LiteLLM DB is
litellm/proxy/auth/handle_jwt.py:2159
↓ 9 callersFunctionsystem_turn
(session: str, mark: JsonValue | None)
tests/integration/routing/_prompt_cache_pin_support.py:67
↓ 9 callersFunctiont_
()
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 9 callersMethodtake_requests
(self, expected_count: int)
tests/rust-python-harness/shared/parity/replay.py:44
↓ 9 callersMethodtarget
(self)
tests/integration/providers/_api_base_trailing_slash_support.py:35
↓ 9 callersFunctionteam_grants
( team_object: LiteLLM_TeamTable | None, team_membership: LiteLLM_TeamMembership | None, user_id:
litellm/proxy/auth/team_grants.py:177
↓ 9 callersMethodterminate
(self)
litellm/proxy/mcp_server/server.py:737
↓ 9 callersMethodtoBaseURI
(s)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 9 callersMethodtoString
()
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 9 callersFunctiontoken_endpoint
Accept the authorization code from client and exchange it for OAuth token. Supports PKCE flow by forwarding code_verifier to upstream provide
litellm/proxy/mcp_server/discoverable_endpoints.py:2446
↓ 9 callersFunctiontool_params_without_hosted_mcp
( params: _ParamsT, model: str, route: str, drop_params: bool, runs_hosted_mcp: Callable[[], bool] | None
litellm/litellm_core_utils/tool_fields.py:301
↓ 9 callersFunctiontool_use_block
( identity: str, *, name: str = "Bash", arguments: Mapping[str, JsonValue] | None = None )
tests/integration/_support/safeguards_rig.py:638
↓ 9 callersFunctiontool_with_sanitized_parameters
Run the tool's JSON schema through ``sanitize``: ``function.parameters`` on an OpenAI tool, ``input_schema`` on an Anthropic one. The same object
litellm/litellm_core_utils/prompt_templates/json_schema_utils.py:451
↓ 9 callersMethodtrackProp
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
↓ 9 callersMethodtrackResult
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
↓ 9 callersFunctiontranscribe_unpriceable_request_reason
( operation: str, request_body: Mapping[str, object], cost_per_second: float | None, )
litellm/proxy/pass_through_endpoints/llm_provider_handlers/transcribe_passthrough_logging_handler.py:163
← previousnext →8,301–8,400 of 202,479, ranked by callers