MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 50 callersFunction_core
gpt (a, b), emb (e), the openai/* wildcard (w), team TA's ``pub`` (t1) and ``openai/*`` (tw), team TB's ``pub`` (t2). Aliases: fast -> gpt, h
tests/unit/test_router/core/test_registry_characterization_lookups.py:16
↓ 50 callersFunction_proxy_attrs_for_centralized_checks
Build the minimal proxy_server module attributes that _run_centralized_common_checks reads. ``master_key`` defaults to a non-None value becau
tests/unit/proxy/auth/test_user_api_key_auth_request_flow.py:4832
↓ 50 callersFunction_success_kwargs
( request_id="req-1", api_key_hash="key-hash", request_metadata=None, call_type="acompletion",
tests/unit/integrations/test_shadow_eval_logger.py:265
↓ 50 callersMethodanswer
(self)
tests/integration/_support/idle_gateway.py:126
↓ 50 callersFunctionappend
(record: BaselineAccountingRecord)
litellm/proxy/db/baseline_accounting.py:651
↓ 50 callersMethodasync_set_cache
(self, *args, **kwargs)
tests/unit/proxy/auth/test_resolvers_store.py:30
↓ 50 callersMethodchat
(self, model: str, *, key: str | None = None, text: str = "integration control")
tests/integration/_support/client.py:153
↓ 50 callersFunctioncreate_response
Create streaming response, checking if the first chunk is an error. If the first chunk is an error, return a standard JSON error response.
litellm/proxy/common_request_processing.py:1460
↓ 50 callersMethodexecute
( &self, request: LiteLLMOcrRequest, interceptors: &impl Interceptors<Error>,
litellm-rust/crates/inference-ocr/src/lib.rs:31
↓ 50 callersFunctioninit_guardrails_v2
( all_guardrails: list[dict], config_file_path: str | None = None, llm_router: Router | None = Non
litellm/proxy/guardrails/init_guardrails.py:20
↓ 50 callersMethoditems
(self)
tests/unit/proxy/pass_through_endpoints/test_pass_through_endpoints.py:12233
↓ 50 callersMethodread
(self)
tests/integration/providers/test_provider_lookup_status.py:157
↓ 50 callersMethodrerank
POST /v1/rerank (Cohere-format). No official OpenAI/Anthropic SDK covers this route, so it stays on the shared typed transport.
tests/e2e/proxy_client.py:1072
↓ 50 callersMethodrespond
(self, request: httpx2.Request)
tests/unit/proxy/mcp_server/test_mcp_server_manager.py:16524
↓ 50 callersFunctionscripted_peer
Raw JSON-RPC peer for shapes the SDK server cannot produce: half-written bodies, stalls, wire errors.
tests/integration/_support/mcp.py:341
↓ 50 callersMethodset_dynamic
(self, **changes: object)
tests/unit/test_router/core/scripted_host.py:300
↓ 50 callersMethodtransform_embedding_request
( self, model: str, input: AllEmbeddingInputValues, optional_params: dict,
litellm/llms/oci/embed/transformation.py:185
↓ 49 callersFunctionC
litellm/proxy/_experimental/out/_next/static/chunks/0r8bcbns1_gw5.js:1
↓ 49 callersFunction_codes
(tmp_path, source)
tests/unit/test_check_test_quality.py:30
↓ 49 callersFunction_dispatch_client_http
(ctx: CompletionDispatchContext)
litellm/main.py:1459
↓ 49 callersFunction_fallback_deployment
(scenario: Scenario, api_base: str, model: str = _OPUS)
tests/integration/providers/test_bedrock_mantle_count_tokens_wire.py:283
↓ 49 callersFunction_install_litellm_config
Ensure mock_prisma.db.litellm_config exists with async methods (the conftest only stubs ``litellm_configtable`` — this is a different table).
tests/unit/proxy/proxy_server/test_routes_config.py:42
↓ 49 callersFunction_json_object
(body: bytes)
tests/integration/observability/_openinference_support.py:75
↓ 49 callersFunction_kinds
(tmp_path: Path, source: str)
tests/unit/test_check_unbounded_in_lists.py:27
↓ 49 callersMethodasync_log_success_event
(self, kwargs, response_obj, start_time, end_time)
tests/unit/integrations/test_s3_v2.py:5918
↓ 49 callersMethodbind
(self, counter_keys: frozenset[str])
litellm/proxy/spend_tracking/spend_counter_batch.py:71
↓ 49 callersFunctionbuild
Build a span and read it back as the JSON the intake receives, not as Python objects.
tests/unit/integrations/datadog/test_datadog_llm_obs.py:87
↓ 49 callersMethodcall
()
tests/unit/proxy/test_common_request_processing.py:11968
↓ 49 callersFunctioncollect_bytes
(stream: object)
tests/unit/test_router/core/streaming_helpers.py:328
↓ 49 callersFunctioncomplete
(core: RouterCore, model: str = "a", **request: object)
tests/unit/test_router/core/test_fallbacks.py:40
↓ 49 callersFunctionenvelope_keys_from_master_key
Derive the envelope signing and encryption keys from the proxy master key. A memory-hard scrypt KDF (RFC 7914) over two distinct domain-label sal
litellm/proxy/mcp_server/outbound_credentials/bridge_credentials.py:54
↓ 49 callersFunctionfinal_text
(marker: str)
tests/integration/providers/_cache_control_marks_support.py:57
↓ 49 callersMethodget_bedrock_route
Get the bedrock route for the given model. Grok and GPT 5.6 and newer go to bedrock-runtime's native OpenAI Chat Completions by
litellm/llms/bedrock/common_utils.py:2053
↓ 49 callersFunctionget_hidden_params
(obj: object)
litellm/litellm_core_utils/hidden_params.py:125
↓ 49 callersFunctionhook
(self: object, data: dict[str, str], call_type: str)
tests/unit/test_internal_context.py:162
↓ 49 callersMethodinitialize_guardrail
Initialize a guardrail from a dictionary and add it to the litellm callback manager Returns a Guardrail object if the guardrail is i
litellm/proxy/guardrails/guardrail_registry.py:740
↓ 49 callersFunctioninstall_adapter
Register a FakeAdapter subclass for every harness and fake the endpoint.
tests/unit/harness/core_fakes.py:212
↓ 49 callersFunctionredact_string
(value: str)
litellm/_logging.py:83
↓ 49 callersFunctionresponses_stream
``completed_response`` is the terminal event the stream latched before failing, if any. ``logged`` logs through the host when the stream ends, wit
tests/unit/test_router/core/streaming_helpers.py:403
↓ 49 callersFunctionrun_centralized_common_checks
Run ``common_checks`` once at the ``user_api_key_auth`` wrapper boundary, regardless of which ``_user_api_key_auth_builder`` path returned. Th
litellm/proxy/auth/user_api_key_auth.py:3811
↓ 49 callersFunctionserializers_value
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 49 callersMethodstart
(self)
tests/unit/proxy/test_proxy_utils.py:2765
↓ 49 callersFunctiontimeout
(name: &str, config: &McpServer)
litellm-rust/crates/gateway-mcp/src/configured.rs:39
↓ 49 callersMethodupdate
(self, *, where: Mapping[str, object], data: Mapping[str, object])
litellm/repositories/prisma_protocols.py:138
↓ 49 callersMethodwait_closed
(self)
tests/unit/proxy/mcp_server/test_mcp_server_manager.py:19755
↓ 48 callersFunction_make_response
(body: dict)
tests/unit/proxy/guardrails/guardrail_hooks/test_singulr.py:50
↓ 48 callersFunction_member_resolution_prisma
Prisma mock where only the given ids resolve to a user row / team row. ``teams`` are teams a SCIM group write created, so they carry provenance;
tests/unit/proxy/management_endpoints/scim/test_scim_v2_endpoints.py:4233
↓ 48 callersFunction_router
One mock router serving the shadow call first, the judge call second, told apart by the internal-origin stamp rather than the model, since a rever
tests/unit/integrations/test_shadow_eval_logger.py:108
↓ 48 callersFunction_spend_counters
(monkeypatch: pytest.MonkeyPatch, **fake: object)
tests/unit/proxy/proxy_server/test_spend_counters.py:52
↓ 48 callersFunction_vertex_deployment
( gateway: Gateway, scenario: Scenario, api_base: str, backend: str, **litellm_params: JsonValue )
tests/integration/providers/test_vertex_anthropic_structured_output_wire.py:795
↓ 48 callersMethodaread
(self)
tests/unit/a2a_protocol/providers/watsonx_orchestrate/test_watsonx_orchestrate_transformation.py:78
↓ 48 callersFunctionattempt
The response, or the error the core raised.
tests/unit/test_router/core/test_retries_cooldowns.py:70
↓ 48 callersFunctionimport_module
(_name: str, _package: str | None = None)
tests/rust-python-harness/strategies/e2e_parity/test_runner.py:32
↓ 48 callersMethodmake_bedrock_api_request
Dispatch to the configured Bedrock guardrail API. ``checks`` selects the resource-less, detect-only InvokeGuardrailChecks API; otherw
litellm/proxy/guardrails/guardrail_hooks/bedrock_guardrails.py:919
↓ 48 callersFunctionmessage_response
()
litellm-rust/crates/inference-messages/tests/messages/main.rs:68
↓ 48 callersMethodnew
(cls)
tests/integration/_support/prometheus_series.py:52
↓ 48 callersFunctionocr_request
A request for an inline PDF, authenticated with `test-key`.
litellm-rust/crates/inference-ocr/tests/ocr/main.rs:94
↓ 48 callersMethodoverride
(self, value: BindingT | None)
litellm/rust_bridge/bindings.py:37
↓ 48 callersFunctionraw_upstream
An upstream that hands each accepted connection to `handle` as raw bytes, for behavior a well-behaved HTTP server would never produce.
litellm-rust/crates/gateway-forward/tests/support/mod.rs:131
↓ 48 callersFunctionrecording_service
()
tests/test_litellm_rust/support/recording_server.py:64
↓ 48 callersFunctionrequest
()
litellm-rust/crates/router-core/tests/call_loop.rs:165
↓ 48 callersMethodreverse
(self)
litellm/router_rust/surface/indexes.py:123
↓ 48 callersMethodset_cache
(self, key, value, **kwargs)
tests/unit/caching/test_caching.py:556
↓ 48 callersFunctiontoken_response
(token: str = "sk-ant-oat01-minted", expires_in: int | None = 3600)
tests/unit/llms/base_llm/auth/test_token_exchange.py:167
↓ 47 callersFunctionY
(e,r)
litellm/proxy/_experimental/out/_next/static/chunks/1a2lzi3ursud4.js:1
↓ 47 callersFunctionassert_one_request
(received: tuple[Request, ...], target: str, marker: str, keyed: bool = True)
tests/integration/providers/_api_base_trailing_slash_support.py:281
↓ 47 callersFunctionassert_served_natively
(server: RecordingServer, *responses: object, count: int = 1)
tests/test_litellm_rust/support/served.py:7
↓ 47 callersMethodcreate
(self, *, data: Mapping[str, object])
litellm/llms/base_llm/managed_resources/base_managed_resource.py:58
↓ 47 callersMethoddelete_key
(self, key: str)
tests/e2e/quota_management/budgets/budget_client.py:258
↓ 47 callersMethodexists
(self, *names: str)
litellm/proxy/mcp_server/outbound_credentials/redis_distributed_lock.py:43
↓ 47 callersFunctionformatNumberWithCommas
( value: number | null | undefined, decimals: number = 0, abbreviate: boolean = false, showZero: boole
ui/litellm-dashboard/src/utils/dataUtils.ts:16
↓ 47 callersFunctionget_or_create_request_stash
()
litellm/proxy/hooks/parallel_request_limiter_v3.py:597
↓ 47 callersFunctionget_policy_registry
Get the global PolicyRegistry singleton. Returns: The global PolicyRegistry instance
litellm/proxy/policy_engine/policy_registry.py:1084
↓ 47 callersMethodget_token
( self, *_: str, claims: str | None = None, tenant_id: str | None = None,
tests/unit/secret_managers/test_get_azure_ad_token_provider.py:631
↓ 47 callersFunctiongithub_http
Swap both module-level HTTP clients for httpx MockTransport versions and collect every request, so tests assert on the real HTTP edge (URL, header
tests/unit/llms/github_copilot/test_per_user_auth.py:73
↓ 47 callersMethodpoll_logs_for_key
( self, key: str, *, min_rows: int = 1, predicate: Callable[[list[Spen
tests/e2e/quota_management/spend_tracking/spend_e2e_client.py:312
↓ 47 callersMethodreceived
(self, request_nonce: str)
tests/integration/messages_endpoint/test_beta_gated_fields_wire.py:121
↓ 47 callersMethodset_attributes
(self, span: Span, kwargs, response_obj: Any | None)
litellm/integrations/opentelemetry.py:2442
↓ 47 callersMethodset_cache
( &self, key: impl Into<String>, value: V, ttl: Option<Duration>, )
litellm-rust/crates/cache-memory/src/cache.rs:96
↓ 47 callersFunctionstream_item_field
(item: object, field: str)
litellm/llms/base_llm/guardrail_translation/utils.py:135
↓ 47 callersFunctiontoken_counter
Args: request: TokenCountRequest call_endpoint: bool - When set to "True" it will call the token counting endpoint - e.g Anthropi
litellm/proxy/proxy_server.py:15972
↓ 47 callersMethodtranslate_anthropic_messages_to_openai
( self, messages: list[AllAnthropicPassThroughMessageValues], model: str | None = None
litellm/llms/anthropic/messages/adapters/transformation.py:595
↓ 47 callersFunctionuiHref
(routeSegment: string)
ui/litellm-dashboard/src/utils/uiHref.ts:14
↓ 47 callersMethodupdate_settings
(self, **kwargs: object)
litellm/router_rust/core.py:2302
↓ 47 callersMethodvalueOf
()
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 46 callersFunctionD
litellm/proxy/_experimental/out/_next/static/chunks/37cr8te-omsl0.js:1
↓ 46 callersFunctionN
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/0r8bcbns1_gw5.js:1
↓ 46 callersFunction_apply_inputs
(messages: list)
tests/unit/proxy/guardrails/guardrail_hooks/test_compresr.py:178
↓ 46 callersFunction_azure_texts
(azure: Wire)
tests/integration/observability/test_azure_content_safety_dispatch.py:1369
↓ 46 callersMethodauth_builder
( api_key: str, jwt_handler: JWTHandler, request_data: dict, general_settings:
litellm/proxy/auth/handle_jwt.py:2497
↓ 46 callersFunctionbind
( legacy: inspect.Signature, args: tuple[object, ...], kwargs: Mapping[str, object] )
litellm/rust_bridge/public_call.py:16
↓ 46 callersMethodcall
Select, run the attempt, and retry or fall back until it succeeds or the policy is exhausted.
litellm-rust/crates/router-core/src/lib.rs:344
↓ 46 callersFunctioncurrent
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 46 callersMethodcurrent
(self)
litellm/proxy/mcp_server/catalog.py:131
↓ 46 callersMethoddone
()
litellm/proxy/_experimental/out/_next/static/chunks/2hk8n3azev--d.js:11
↓ 46 callersMethoddone
(self)
litellm/caching/redis_batch.py:291
↓ 46 callersMethodenter
(e,t,n,s)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 46 callersMethodfind_many
( self, where: Mapping[str, object] | None = None, include: Mapping[str, b
litellm/proxy/management_endpoints/customer_endpoints.py:82
↓ 46 callersMethodfrom_env
(cls, environ: Mapping[str, str] = os.environ)
litellm/proxy/lens/adapter.py:53
↓ 46 callersMethodgenerate_key
( self, *, models: list[str] | None = None, max_budget: float | None = None,
tests/e2e/quota_management/budgets/budget_client.py:228
← previousnext →901–1,000 of 202,479, ranked by callers