MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 7 callersFunctionpool_config
(options: ReloadOptions, workers: int, worker_overrides: Mapping[str, object] | None = None)
litellm/proxy/dev_reload.py:92
↓ 7 callersMethodpop
(&self, key: &str, now: f64)
litellm-rust/crates/cache-disk/src/sqlite.rs:442
↓ 7 callersMethodposBeforeChild
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 7 callersFunctionpost_call_processing
( original_response, model, optional_params: dict | None, original_function, rules_obj, )
litellm/utils.py:1954
↓ 7 callersFunctionposted_inputs
(wire: Wire)
tests/integration/_support/responses_replay.py:209
↓ 7 callersMethodpre_seed_litellm_metadata_for_route
Pre-seed ``litellm_metadata`` for routes that track tags there. Routes in ``LITELLM_METADATA_ROUTES`` (e.g. Bedrock, ``/v1/messages``,
litellm/proxy/litellm_pre_call_utils.py:2370
↓ 7 callersFunctionprefetch_identity_keys
Warm the entries auth reads before it knows the key's owners (the key object, the end user and the two registries) in one read on the request's na
litellm/proxy/auth/auth_object_prefetch.py:325
↓ 7 callersMethodprepare
( agent: &impl Configure, installed: &Installed, settings: Settings, home: imp
litellm-rust/crates/testkit/src/session.rs:23
↓ 7 callersFunctionprepare_agent_invocation
( auth: UserAPIKeyAuth, target_name: str, store: AgentIdentityStore | None, *, billable: bool = True )
litellm/proxy/agent_endpoints/auth/managed_authorization.py:238
↓ 7 callersFunctionprepare_context_managed_request
Apply client compaction history, then optional context_management polyfill.
litellm/llms/anthropic/messages/adapters/handler.py:295
↓ 7 callersFunctionprint_verbose
(print_statement)
litellm/proxy/example_config_yaml/custom_callbacks.py:22
↓ 7 callersFunctionprobe
()
tests/unit/caching/test_redis_cache.py:1601
↓ 7 callersFunctionprocess_is_gone
(pid: int, within_seconds: float)
tests/_process_helpers.py:51
↓ 7 callersMethodprocess_pre_call_hook_response
(self, response, data, call_type)
litellm/proxy/utils.py:2582
↓ 7 callersFunctionprovider_edge_api_base
The api_base a suite gives an edge-wired deployment: None in live mode (the deployment keeps its real provider api_base) and the process-wide edge
tests/e2e/provider_edge.py:1253
↓ 7 callersFunctionpurge_user_oauth_credentials_for_server
Delete every stored per-user OAuth token for a server and invalidate each user's cached token everywhere it can be served from (the legacy per-use
litellm/proxy/mcp_server/db.py:2005
↓ 7 callersMethodquery_raw
(self, query: str, *args: str | Sequence[str])
litellm/proxy/management_helpers/access_group_key_sync.py:50
↓ 7 callersFunctionr
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 7 callersFunctionraise_if_reload_degraded_serving
The caller-visible error this pod's model-write endpoints owe their caller when the model they wrote is not being served after the reload they tri
litellm/proxy/management_endpoints/model_management_endpoints.py:3990
↓ 7 callersFunctionrb
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 7 callersFunctionread
()
tests/integration/cost_calculation/conftest.py:110
↓ 7 callersMethodread
`("redis" | "memory", value)`: where the answer came from, a Redis miss included.
litellm-rust/crates/python-bridge/src/redis/store/spend_counters.rs:77
↓ 7 callersMethodread_or_unset
Reads the accessor, or `None` when the litellm package is not installed (a bare extension module), meaning there are no configured values.
litellm-rust/crates/python-bridge/src/python_settings.rs:46
↓ 7 callersFunctionrealtime_guarded_item
(client_event: Mapping[str, object])
litellm/litellm_core_utils/realtime_guardrails.py:100
↓ 7 callersFunctionreconcile_budget_reservation
Settle every reserved counter on ``actual_cost``. With ``apply_consistent`` False the adjustments for counters that still hold the reservation are
litellm/proxy/spend_tracking/budget_reservation.py:353
↓ 7 callersFunctionrecord
(db: Prisma)
tests/integration/spend/test_baseline_accounting.py:56
↓ 7 callersMethodrecord
(self, delivery: Delivery)
tests/integration/_support/mail.py:39
↓ 7 callersMethodrecord
(&self, event: String)
litellm-rust/crates/cache/tests/dual.rs:412
↓ 7 callersFunctionrecord_shadow_eval_funnel_event
Count one skipped request for one job leg; synchronous so the hook's read-modify- write cannot interleave with the flush's snapshot on the shared
litellm/proxy/db/shadow_eval_funnel.py:41
↓ 7 callersFunctionrecord_stream
(root: Path, *, abort_after: int | None = None)
tests/e2e_harness/test_provider_edge.py:978
↓ 7 callersFunctionrecover_key_owner_from_daily_spend
( prisma_client: PrismaClient, keys: AbstractSet[str], )
litellm/proxy/spend_tracking/key_metadata_recovery.py:227
↓ 7 callersMethodrecv
(self, decode: Optional[bool] = None)
tests/_ws_vcr.py:268
↓ 7 callersMethodredact_headers_for_logging
Create a copy of headers with sensitive values redacted for safe logging. Shows first 8 characters of sensitive values for debugging
litellm/llms/databricks/common_utils.py:152
↓ 7 callersFunctionredact_internal_details
(value: str)
litellm/litellm_core_utils/secret_redaction.py:124
↓ 7 callersFunctionredact_sensitive_agent_litellm_params
Replace credential-bearing values in an agent's litellm_params with ``REDACTED_BY_LITELM_STRING`` while preserving non-secret keys (``model``
litellm/proxy/agent_endpoints/agent_registry.py:211
↓ 7 callersMethodredact_tool_payload
Redact secrets in a tool payload's string values, scanning each value alone and as its JSON lines.
enterprise/litellm_enterprise/enterprise_callbacks/secret_detection.py:667
↓ 7 callersMethodrefetchQueries
(e,t={})
litellm/proxy/_experimental/out/_next/static/chunks/0tx19jeq-wskb.js:2
↓ 7 callersFunctionrefusal_reply_text
(message_or_delta: object)
litellm/llms/anthropic/messages/native/utils.py:169
↓ 7 callersFunctionregister_model
Register new / Override existing models (and their pricing) to specific providers. Provide EITHER a model cost dictionary or a url to a hoste
litellm/utils.py:4404
↓ 7 callersFunctionregister_user_var_server
(scenario: Scenario, peer: McpPeer, *names: str)
tests/integration/mcp/test_mcp_user_env_vars.py:39
↓ 7 callersFunctionregistration
(name: str, **fields: JsonValue)
tests/integration/management/test_claude_code_plugin_installation_preference.py:45
↓ 7 callersMethodreinitialize_guardrail
Force re-initialization of a guardrail even if it exists in memory. Removes old callback from litellm.callbacks and creates fresh ins
litellm/proxy/guardrails/guardrail_registry.py:1109
↓ 7 callersMethodremember
(self, scope: tuple[str, str], cache_key: str, name: str | None, expire_time: object)
litellm/llms/vertex_ai/context_caching/known_cached_contents.py:93
↓ 7 callersFunctionremove
(name: string, id: string)
tests/e2e/ui/tests/usage/usagePage.spec.ts:107
↓ 7 callersMethodremove
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 7 callersMethodremove
(state: &mut CacheState<V>, key: &str)
litellm-rust/crates/cache-memory/src/cache.rs:283
↓ 7 callersMethodremove_callback_from_all_lists
Remove a callback object from every callback list it may have been promoted into, so a re-initialized callback leaves no stale instan
litellm/litellm_core_utils/logging_callback_manager.py:431
↓ 7 callersFunctionremove_file
(directory: &Path, filename: &str)
litellm-rust/crates/cache-disk/src/sqlite.rs:807
↓ 7 callersMethodreplay_model_cost_registrations
Re-register every served deployment's cost-map entries (``replay_model_cost_registrations``).
litellm/router_rust/surface/registry.py:1523
↓ 7 callersMethodreport_failure
The stream failed after it opened: count the failure, cool down by the rules and count one more request (``async_deployment_callback_on_failur
litellm/router_rust/invoke/streaming.py:222
↓ 7 callersFunctionreported_everywhere
(candidates: tuple[Gateway, ...], key: str, model: str, expected: str | None)
tests/integration/management/test_budget_reset_windows.py:210
↓ 7 callersFunctionrequest
()
litellm-rust/crates/cache-response/tests/support/mod.rs:16
↓ 7 callersFunctionrequest
( route: &MessagesRoute, call: MessagesCall, host: &LiveCall, )
litellm-rust/crates/inference-messages/tests/live/support.rs:75
↓ 7 callersFunctionrequest_dispatched_to_pass_through_endpoint
Whether FastAPI resolved this request to a user-defined pass-through handler. Reads the marker set by ``create_pass_through_route`` off the dispa
litellm/proxy/auth/auth_utils.py:2399
↓ 7 callersFunctionrequest_github
( client: httpx.AsyncClient, method: str, path: str, params: Mapping[str, str | int] | None =
litellm/proxy/roi_calculator/github.py:223
↓ 7 callersFunctionrequest_json
(request: Request)
tests/integration/providers/_api_base_trailing_slash_support.py:60
↓ 7 callersFunctionrequest_safeguards_evaluator
( kwargs: Mapping[str, object], messages: Sequence[Mapping[str, object]] )
litellm/llms/anthropic/messages/adapters/handler.py:277
↓ 7 callersMethodrequests
(self)
tests/integration/management/test_credential_federation_values.py:515
↓ 7 callersFunctionrequires_native_compaction_beta
( custom_llm_provider: str, optional_params: Mapping[str, object], messages: Sequence[object], )
litellm/llms/anthropic/common_utils.py:265
↓ 7 callersMethodreset_budget_for_litellm_teams
Resets the budget for all LiteLLM Internal Teams if their budget has expired
litellm/proxy/common_utils/reset_budget_job.py:1355
↓ 7 callersFunctionreset_mcp_message_trace_carrier
(token: "Token[Mapping[str, str] | None]")
litellm/integrations/otel/plumbing/context.py:124
↓ 7 callersMethodresolve
(agent_id: str)
tests/unit/proxy/agent_endpoints/auth/test_agent_permission_handler.py:311
↓ 7 callersFunctionresolveLogoSrc
(value: string | null | undefined, root: string = serverRootPath)
ui/litellm-dashboard/src/lib/assetPaths.ts:25
↓ 7 callersFunctionresolveSchemaProperty
(prop: InputSchemaProperty)
ui/litellm-dashboard/src/app/(dashboard)/mcp-servers/_components/toolCallArguments.ts:20
↓ 7 callersFunctionresolve_baggage_tag_mappings
Exact mappings win over the any-value fallback for the same baggage key.
litellm/proxy/common_utils/request_tag_mappings.py:34
↓ 7 callersFunctionresolve_baseline
The derived baseline for a router whose hardest tier offers ``group_names``. Holds no cache of its own; each pricing pass walks the pool, so the
litellm/router_strategy/savings_baseline.py:147
↓ 7 callersFunctionresolve_bedrock_request_metadata
Resolve the ``requestMetadata`` pairs to send to Bedrock, or ``None`` when nothing survives Bedrock's constraints. The result is a plain dict
litellm/llms/bedrock/request_metadata.py:131
↓ 7 callersMethodresolve_connection
(self, api_base: str | None, api_key: str | None)
litellm/llms/laya/decisions/transformation.py:13
↓ 7 callersFunctionresolve_dashscope_family_rerank_api_base
(api_base: str | None, env_var: str, default_rerank_base: str)
litellm/llms/dashscope/common_utils.py:169
↓ 7 callersFunctionresolve_explicit_cache_read_cost_rate
( model_info: ModelInfo, usage: Usage, current_time: datetime | None, service_tier: str | None
litellm/litellm_core_utils/llm_cost_calc/utils.py:582
↓ 7 callersFunctionresolve_fireworks_resource_name
(model: str)
litellm/llms/fireworks_ai/common_utils.py:93
↓ 7 callersFunctionresolve_owned_read_scope
Resolve own-user and permitted-team reads, falling back to own-user on lookup failure.
litellm/proxy/auth/authorization.py:26
↓ 7 callersFunctionresolve_session_signing_keys
Turn the operator's ``mcp_session_token_signing`` setting into signing key material. ``None`` (the setting absent) keeps the backward-compatible
litellm/proxy/mcp_server/outbound_credentials/session_credentials.py:124
↓ 7 callersFunctionresolve_upstream_resource
Resolve the RFC 8707 ``resource`` value this server's upstream OAuth legs must carry. The MCP authorization spec requires an MCP client to send `
litellm/proxy/mcp_server/oauth_utils.py:785
↓ 7 callersFunctionresolve_vertex_location_for_cost
The Vertex AI location a request was served from, resolved the same way dispatch resolves it, so regional deployments price with the regi
litellm/litellm_core_utils/litellm_logging.py:561
↓ 7 callersFunctionresolver
()
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 7 callersFunctionresourceLiteLLMTeamCreate
(d *schema.ResourceData, m interface{})
terraform/provider/litellm/resource_team.go:177
↓ 7 callersFunctionrespond_in_order
Scripts responses on a started server, for responses that need its address.
litellm-rust/crates/inference-ocr/tests/support/mod.rs:47
↓ 7 callersMethodresponse_object_includes_web_search_call
Check if the response object includes a web search call. This covers: - Chat Completion Response (ModelResponse) - R
litellm/litellm_core_utils/llm_cost_calc/tool_call_cost_tracking.py:533
↓ 7 callersFunctionretain_sso_identity_assertion_for_ema
The SSO-callback hook: a no-op unless there is material AND an EMA server is registered. A store failure is logged and swallowed because the login
litellm/proxy/mcp_server/outbound_credentials/sso_assertion_store.py:508
↓ 7 callersFunctionretried_ok
()
tests/unit/test_router/core/test_streaming_characterization_responses.py:506
↓ 7 callersFunctionrewrite_batch_input_file
Re-emit the file with redacted records rewritten and dropped records left out. Untouched records are copied through as written rather than r
litellm/proxy/openai_files_endpoints/batch_guardrails.py:535
↓ 7 callersFunctionriva_server
( words: tuple[Word, ...], *, offline_only: bool = False, stall: bool = False, max_receive
tests/integration/_support/riva.py:131
↓ 7 callersFunctionroute_cache_model
( params: LiteLLMParamsBody, base_for: Callable[[str], str | None], *, enabled: bool, mode: ModelMode | No
tests/e2e/provider_cache_routing.py:72
↓ 7 callersFunctionrouted_attempt
()
litellm/litellm_core_utils/attempt_settlement.py:45
↓ 7 callersFunctionrouter_model_routes
(llm_router: "Router | None", team_id: str | None)
litellm/proxy/management_helpers/model_quota_policy.py:207
↓ 7 callersMethodrpc
(self, method: str, params: JsonRpc | None = None)
tests/integration/_support/mcp.py:685
↓ 7 callersMethodrun
( self, user_id: str, server_id: str, refresh: Callable[[], Awaitable[OAuthTok
litellm/proxy/mcp_server/outbound_credentials/redis_refresh_coordinator.py:85
↓ 7 callersMethodrun
(self, cmd, *, env=None, cwd=None, timeout=None)
tests/unit/llms/codex/harness/test_transformation.py:116
↓ 7 callersFunctionrunKeyUpdate
(t *testing.T, p *keyRecoveryProxy, d *schema.ResourceData)
terraform/provider/litellm/resource_key_test.go:1341
↓ 7 callersFunctionrun_call
( py: Python<'_>, start: impl FnOnce( Python<'_>, &Bound<'_, PyDict>, <H::Prot
litellm-rust/crates/host-python/src/driver.rs:587
↓ 7 callersFunctionrun_len
Byte length of the leading run of `run_class` characters.
litellm-rust/crates/token-counter-fast/src/unicode_classes.rs:111
↓ 7 callersFunctionrun_prisma
Run one Prisma CLI command in its own process group, bounded by ``timeout``. Raises ``subprocess.TimeoutExpired`` once the budget is spent, after
litellm-proxy-extras/litellm_proxy_extras/prisma_toolchain.py:217
↓ 7 callersMethodrun_scheduled_deprecation_check
Poll every pass for a loaded router, the alert being on, and no alert in the last day, then alert A pass that could not alert (no router yet,
litellm/integrations/SlackAlerting/slack_alerting.py:1177
↓ 7 callersFunctionrun_scripted
( py: Python<'_>, machine: MachineFactory, op: OpScript, script: HookScript, asynchronous:
litellm-rust/crates/host-python/tests/support/mod.rs:506
↓ 7 callersFunctionrun_suites
( cases: Sequence[HarnessCase], repo_root: Path, on_update: UpdateCallback, runner_args: Seque
tests/rust-python-harness/shared/unit_runners/suite_runner.py:32
↓ 7 callersMethodrun_sync
(self)
litellm/router_rust/core.py:4225
↓ 7 callersMethodruntime
( # noqa: ANN201 # boto3 clients are untyped self, monkeypatch: pytest.MonkeyPatch, extra_headers: M
tests/integration/providers/test_bedrock_passthrough_guardrail_config_wire.py:273
↓ 7 callersFunctionsA
litellm/proxy/_experimental/out/_next/static/chunks/269vry4mb17-_.js:3
↓ 7 callersFunctionsa
litellm/proxy/_experimental/out/_next/static/chunks/2k-fkubkt19cc.js:8
← previousnext →11,201–11,300 of 202,479, ranked by callers