MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 12 callersMethod_build_kwargs
(self, *, exception: bool = False)
tests/unit/integrations/test_opentelemetry.py:5216
↓ 12 callersFunction_build_prisma_with_keys
(user_keys, mock_user=None, updated_user=None)
tests/unit/proxy/management_endpoints/scim/test_scim_key_deactivation.py:39
↓ 12 callersFunction_build_simple_text_stream
Produce SSE bytes for a simple text response (no tool calls).
tests/unit/llms/anthropic/messages/native/test_agentic_streaming_iterator.py:39
↓ 12 callersFunction_build_tool_use_stream
Produce SSE bytes for a response with a tool_use block.
tests/unit/llms/anthropic/messages/native/test_agentic_streaming_iterator.py:103
↓ 12 callersFunction_cache
A worker of its own (empty memory) on ``redis``'s Redis, or a worker without Redis.
tests/unit/proxy/auth/test_auth_object_prefetch.py:88
↓ 12 callersFunction_call
One call on ``op``: an ``a``-prefixed op is awaited, the other one called.
tests/unit/test_router/core/test_parity_decisions_routing.py:53
↓ 12 callersMethod_call
(self, request_body)
tests/unit/proxy/auth/test_auth_checks_object_access_and_lookup.py:5929
↓ 12 callersFunction_call_tool_result
(is_error: bool, text: str)
tests/unit/proxy/mcp_server/test_mcp_server_tool_calls_and_headers.py:9735
↓ 12 callersFunction_caller
A team, an internal user on it and that user's key on the team, allowed ``models``.
tests/integration/security/test_stored_config_slots.py:106
↓ 12 callersFunction_callers
(db, actors, prefix: str)
tests/integration/management/test_team_update.py:54
↓ 12 callersFunction_chat
One chat request; returns the response and the spend-log request id.
tests/integration/security/test_stored_config_slots.py:146
↓ 12 callersFunction_chat
( gateway: Gateway, body: dict[str, JsonValue], *, key: str | None = None, headers: dict[s
tests/integration/routing/test_fallback_retry_header_contracts.py:41
↓ 12 callersFunction_chat
(*blocks: object, role: str = "user")
tests/unit/llms/bedrock/test_guardrail_attachments.py:22
↓ 12 callersFunction_check_passthrough_upstream_auth
Probe pass-through and legacy delegate upstream servers in parallel before the MCP session starts. Only servers the caller's key is already a
litellm/proxy/mcp_server/server.py:2212
↓ 12 callersFunction_clear_prometheus_registry
()
tests/unit/integrations/test_prometheus_labels.py:17
↓ 12 callersFunction_client_with_session
( app: Starlette, admission: Callable[[], AbstractAsyncContextManager[None]] | None = None, )
tests/unit/mcp_client/test_mcp_client.py:4314
↓ 12 callersFunction_coded_failure_response
(status: int, code: str | None, raw_body: str | None = None)
tests/unit/integrations/test_s3_v2.py:3350
↓ 12 callersFunction_collect
(stream: AsyncIterator[bytes])
tests/unit/llms/anthropic/messages/native/test_response_cache.py:68
↓ 12 callersFunction_collect_async
(chunks: List[str])
tests/unit/llms/oci/chat/test_oci_sse_splitter.py:24
↓ 12 callersFunction_collect_calls
Helper: return dict[attr_name] = value of all set_attribute calls.
tests/unit/integrations/arize/test_arize_utils.py:476
↓ 12 callersFunction_collect_sync
(stream: Iterator[str])
tests/unit/llms/oci/chat/test_oci_sse_splitter.py:20
↓ 12 callersMethod_commit_spend_updates_to_db_with_redis
Handler to commit spend updates to Redis and attempt to acquire lock to commit to db This is a v2 scalable approach to first commit
litellm/proxy/db/db_spend_update_writer.py:1789
↓ 12 callersFunction_config
( *, source: IdJagSubjectSource = IdJagSubjectSource.refresh_token, client_id: str = "agent", endpoint: st
tests/unit/proxy/mcp_server/outbound_credentials/test_login_assertion_exchange.py:42
↓ 12 callersFunction_config
(**overrides: Any)
tests/unit/integrations/focus/test_ternary_destination.py:27
↓ 12 callersFunction_configure_saved_agent
(runner: CliRunner, target: Literal["claude", "codex"])
tests/unit/proxy/client/cli/test_configure_commands.py:726
↓ 12 callersFunction_conjuncts
(plan: QueryPlan)
tests/unit/proxy/list_api/test_list_framework.py:133
↓ 12 callersFunction_consumed
(gateway: Gateway, path: str, body: Mapping[str, JsonValue])
tests/integration/providers/test_bedrock_mantle_context_overflow_wire.py:195
↓ 12 callersFunction_container
(container_id: str)
tests/unit/containers/test_container_proxy_ownership.py:34
↓ 12 callersFunction_content_filter
Instantiate ContentFilterGuardrail with a given category.
litellm/proxy/guardrails/guardrail_hooks/litellm_content_filter/guardrail_benchmarks/test_eval.py:243
↓ 12 callersFunction_conversation_span
The exported LLM-call span for ``payload`` with content capture on. ``span_limits`` builds the provider with programmatic limits instead of the e
tests/unit/integrations/otel/test_otel_v2_emitter.py:502
↓ 12 callersFunction_converse_cell
(gateway: Gateway, wire: Wire, body: Mapping[str, JsonValue])
tests/integration/providers/test_bedrock_converse_missing_content_wire.py:277
↓ 12 callersMethod_convert_mcp_to_llm_format
(self, request_obj, kwargs)
tests/unit/proxy/mcp_server/test_mcp_server_manager.py:20474
↓ 12 callersMethod_convert_tools_to_responses_format
Convert chat completion tools to responses API tools format
litellm/completion_extras/litellm_responses_transformation/transformation.py:1904
↓ 12 callersFunction_cooldown_ctx
A RoutingContext double for the fallback cooldown trigger: no deployment, this model info, this cooldown time.
tests/unit/router_utils/test_fallback_event_handlers.py:982
↓ 12 callersMethod_core_get_deployment
(self, model_id: str)
litellm/router_rust/surface/registry.py:1564
↓ 12 callersFunction_count
(gateway: Gateway, model: str, status: int, count_model_status: int | None = None)
tests/integration/providers/test_count_tokens_model_wire.py:275
↓ 12 callersFunction_count_cache_control
Count cache_control breakpoints across messages (message + content level).
tests/unit/integrations/test_anthropic_cache_control_hook.py:1054
↓ 12 callersFunction_crash_lines
(window: str)
tests/integration/authorization/test_websocket_rejection_logging.py:104
↓ 12 callersFunction_customer_sample
(metric_name: str, end_user_id: str)
tests/unit/integrations/test_prometheus_user_team_metrics.py:914
↓ 12 callersFunction_dashboard_sign_in
(proxy: Gateway, graph: Graph, hint: str)
tests/integration/authorization/test_sso_blank_identity.py:428
↓ 12 callersFunction_day
(date: str, model_groups: dict[str, MetricWithMetadata])
tests/unit/proxy/management_endpoints/test_ptu_consumption.py:50
↓ 12 callersFunction_db_health_readiness_check
()
litellm/proxy/health_endpoints/_health_endpoints.py:1964
↓ 12 callersFunction_db_systems
(spans: tuple[Span, ...])
tests/integration/observability/test_otel_excluded_services.py:228
↓ 12 callersFunction_decide
( gateway: Gateway, model: str, questions: Sequence[Mapping[str, JsonValue]] = _QUESTIONS, *,
tests/integration/providers/test_decisions_box_deployment_flag_wire.py:197
↓ 12 callersFunction_delete_credential
(name: str, auth=_as_admin)
tests/unit/proxy/credential_endpoints/test_endpoints.py:75
↓ 12 callersFunction_delta
(index: int, content: str)
tests/integration/routing/test_vllm_passthrough_stream_relay.py:74
↓ 12 callersMethod_delta
(sequence_number, text)
tests/unit/llms/openai/responses/test_openai_responses_guardrail_handler.py:4896
↓ 12 callersFunction_deployment
(scenario: Scenario, wire: Wire, **overrides: JsonValue)
tests/integration/providers/test_bedrock_runtime_chat_completions_sad_wire.py:59
↓ 12 callersFunction_deployment
( scenario: Scenario, wire: Wire, extra: Mapping[str, JsonValue], *, provider: str = "sagemaker_chat" )
tests/integration/providers/test_sagemaker_chat_wire.py:62
↓ 12 callersFunction_deployment
( scenario: Scenario, runtime_port: int, model: str = _SONNET, model_info: Mapping[str, JsonVa
tests/integration/providers/test_count_tokens_model_wire.py:265
↓ 12 callersFunction_deployments_hit
(wire: Wire)
tests/integration/sdk/test_router_sync_stream_fallback_wire.py:59
↓ 12 callersFunction_describe
(received: tuple[Request, ...])
tests/integration/providers/test_device_code_login_guard_wire.py:387
↓ 12 callersFunction_digest
(key: str)
tests/integration/authorization/test_unknown_key_generic_401.py:107
↓ 12 callersFunction_dotted_name
`a.b.c` for an attribute chain rooted in a plain name, else "".
scripts/check_test_quality.py:219
↓ 12 callersMethod_drive
( self, hook_kwargs: Mapping[str, object], response_obj: object, *, gu
tests/unit/integrations/test_shadow_eval_logger.py:333
↓ 12 callersMethod_drop_listed_tools
(self, server_id: str)
litellm/proxy/mcp_server/mcp_server_manager.py:4685
↓ 12 callersFunction_embed
(candidate: Gateway, model: str, **extra: JsonValue)
tests/integration/providers/test_voyage_mongodb_host_wire.py:221
↓ 12 callersFunction_encode_realtime_token_payload
Encode metadata with the upstream ephemeral key so /realtime/calls can route without requiring model as a query param.
litellm/proxy/realtime_endpoints/endpoints.py:264
↓ 12 callersFunction_enforce
( matched: tuple[str, ...], *rows: _MagBudgetRow, spend_by_counter_key: dict[str, float] | None =
tests/unit/proxy/auth/test_model_access_group_budgets.py:339
↓ 12 callersMethod_enforce_batch_file_model_access
Reject the batch if the caller is not authorized for the upload target. For managed files, ``target_model_names`` (from the unified file id)
litellm/proxy/hooks/batch_rate_limiter.py:1062
↓ 12 callersFunction_env
(lock_dir: Path, slots: str)
tests/unit/test_gate_slot_lock.py:47
↓ 12 callersFunction_estimate
(mock_router: MagicMock | None, model: str = AN_ALIAS, **overrides: int)
tests/unit/proxy/management_endpoints/test_cost_tracking_settings.py:886
↓ 12 callersFunction_exchange
( openai_gateway: Gateway, openai_api: ControlPlane, body: Callable[[str, str], Mapping[str, JsonV
tests/integration/providers/test_openai_responses_tool_search_wire.py:161
↓ 12 callersMethod_execute_agentic_loop
Legacy path: execute search + build patch + run follow-up call.
litellm/integrations/websearch_interception/handler.py:1365
↓ 12 callersFunction_expected_upstream_body
(marker: str, **media: JsonValue)
tests/integration/providers/test_cloudflare_clef_omni_media_wire.py:157
↓ 12 callersFunction_expires_at
()
tests/integration/providers/test_realtime_webrtc_http.py:56
↓ 12 callersFunction_extra
(endpoint: Endpoint)
tests/integration/providers/test_gemini_tool_result_part_reference_wire.py:235
↓ 12 callersMethod_extract_all_messages
Extract all messages (request + response) with sequence numbers and timestamps. Processes request messages from StandardLoggingPaylo
litellm/integrations/newrelic/newrelic.py:506
↓ 12 callersFunction_fails
(core: RouterCore, model: str = "a", **request: object)
tests/unit/test_router/core/test_parity_pick_count_and_fallback_log.py:67
↓ 12 callersFunction_fake_redisvl_modules
(semantic_cache_mock: MagicMock, custom_vectorizer_mock: MagicMock)
tests/unit/caching/test_redis_semantic_cache.py:14
↓ 12 callersFunction_fallback_logs
(messages: list[str])
tests/unit/test_router/core/test_parity_pick_count_and_fallback_log.py:63
↓ 12 callersFunction_feature_module
(monkeypatch: pytest.MonkeyPatch, name: str, path: str)
tests/unit/proxy/test__lazy_features.py:51
↓ 12 callersMethod_fetch_single_authorization_server_metadata
( self, issuer_url: str, server_url: str, require_issuer: str | None = None )
litellm/proxy/mcp_server/mcp_server_manager.py:5466
↓ 12 callersFunction_flat
(groups)
tests/unit/llms/openai/responses/test_openai_responses_guardrail_tool_merge.py:19
↓ 12 callersFunction_function_tool_call
(call_id: str | None, name: str, arguments: str)
tests/unit/litellm_core_utils/prompt_templates/test_litellm_core_utils_prompt_templates_factory.py:5250
↓ 12 callersFunction_gateway_url
(gateway: Gateway)
tests/integration/providers/test_realtime_unguarded_voice_turns.py:174
↓ 12 callersFunction_get_allowed_mcp_servers
Return allowed MCP servers for a request after applying filters. Args: user_api_key_auth: The authenticated user's API key info.
litellm/proxy/mcp_server/operations.py:842
↓ 12 callersMethod_get_async_client_session
(self, dynamic_client_session: ClientSession | None = None)
litellm/llms/custom_httpx/aiohttp_handler.py:105
↓ 12 callersMethod_get_azure_config
Get Azure config based on model type. When *base_model* is provided (e.g. ``"azure/gpt-5.2"``), it is used for model-type detection i
litellm/utils.py:10048
↓ 12 callersMethod_get_call_fn
(self)
tests/unit/proxy/mcp_server/test_mcp_tool_search.py:675
↓ 12 callersMethod_get_key_access_group_mcp_server_extras
Resolve the key's unified `access_group_ids` (LiteLLM_AccessGroupTable) to MCP server IDs as additive grants: a group attached to the
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:3513
↓ 12 callersFunction_get_prisma_command
Get the Prisma command to use, bypassing Python wrapper in offline mode.
litellm-proxy-extras/litellm_proxy_extras/utils.py:227
↓ 12 callersFunction_get_prisma_env
Get environment variables for Prisma, handling offline mode if configured.
litellm-proxy-extras/litellm_proxy_extras/utils.py:55
↓ 12 callersFunction_groups
(response)
tests/unit/proxy/public_endpoints/public_v1/test_model_hub.py:65
↓ 12 callersFunction_handle_stream_message
Handle message/stream method via SDK functions. The A2A JSON-RPC binding streams responses as SSE (text/event-stream) with each JSON-RPC obje
litellm/proxy/agent_endpoints/a2a_endpoints.py:547
↓ 12 callersFunction_handler
()
tests/unit/proxy/guardrails/guardrail_hooks/akto/test_akto.py:37
↓ 12 callersFunction_headers
(marker: str, retry_after: str = "7")
tests/integration/sdk/test_stream_open_error_provider_headers_sdk.py:75
↓ 12 callersFunction_health
A breaker no other test shares, so one test opening it cannot fast-fail another.
tests/unit/caching/test_redis_cache.py:656
↓ 12 callersFunction_image_response
(num_images: int = 1)
tests/unit/llms/fal_ai/test_cost_calculator.py:20
↓ 12 callersFunction_index_validity
(schema: str)
tests/integration/database/migrations/test_invalid_index_repair.py:36
↓ 12 callersMethod_init_litellm_callbacks
(self, llm_router: Router | None = None)
litellm/proxy/utils.py:2172
↓ 12 callersMethod_inject
(self, messages, kwargs, system="sys", tools=None)
tests/unit/integrations/test_anthropic_cache_control_hook.py:3209
↓ 12 callersMethod_inject
(messages, system, kwargs, model="openai/gpt-5.6", custom_llm_provider=None)
tests/unit/integrations/test_anthropic_cache_control_hook.py:3567
↓ 12 callersFunction_inputs
(messages: list[dict[str, object]])
tests/unit/proxy/guardrails/guardrail_hooks/test_typesafe.py:99
↓ 12 callersFunction_install_login_mocks
Patch authenticate_user + create_ui_token_object at their import paths. Both /login, /v2/login and /v3/login do a *local* (in-function) import of
tests/unit/proxy/proxy_server/test_routes_login_sso.py:28
↓ 12 callersFunction_install_real_auth
Run the real authenticate_user so the throttle inside it is exercised. prisma_client stays None, so every guess falls through to the credential r
tests/unit/proxy/proxy_server/test_routes_login_sso.py:514
↓ 12 callersFunction_invoke
Run invoke_agent_a2a with standard patches applied.
tests/unit/proxy/agent_endpoints/test_agent_headers.py:108
↓ 12 callersMethod_is_already_running_as_role
Check if the current environment is already running as the target IAM role. This handles multiple AWS environments: - IRSA (
litellm/llms/bedrock/base_aws_llm.py:888
↓ 12 callersMethod_is_idempotent_error
Check if the error message indicates an idempotent operation error. Idempotent errors (like "column already exists") mean the migrat
litellm-proxy-extras/litellm_proxy_extras/utils.py:458
↓ 12 callersFunction_is_mapping
(value: object)
litellm/llms/bedrock/guardrail_attachments.py:159
← previousnext →5,201–5,300 of 202,479, ranked by callers