MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 7 callersFunction_carry_guardrail_logging_info
Copy guardrail logging entries from ``guardrail_data`` onto ``request_data``. Post-call guardrails run against a throwaway ``hook_data`` dict (it
litellm/proxy/pass_through_endpoints/pass_through_endpoints.py:881
↓ 7 callersFunction_chain_allows_fail_open
( healthy_deployments: _DeploymentPool, request_deployments: Iterable[_DeploymentLike | _DeploymentWit
litellm/router_strategy/tag_based_routing.py:289
↓ 7 callersFunction_change
(*model_ids: str)
tests/unit/proxy/common_utils/test_config_sync_pubsub.py:185
↓ 7 callersFunction_chat
(deployment: _Deployment, **extra: JsonValue)
tests/integration/observability/test_prometheus_deployment_failure_api_provider.py:255
↓ 7 callersFunction_chat
(gateway: Gateway, model: str, token: str, prompt: str)
tests/integration/authorization/test_jwt_key_mapping_lifecycle.py:300
↓ 7 callersFunction_chat
(gateway: Gateway, body: Mapping[str, JsonValue])
tests/integration/streaming/test_responses_bridge_stream_errors.py:101
↓ 7 callersFunction_chat
(proxy: Gateway, model: str, key: str, text: str)
tests/integration/management/test_team_member_lifecycle.py:156
↓ 7 callersFunction_chat
(gateway: Gateway, model: str, extra: Mapping[str, JsonValue] = NO_EXTRA)
tests/integration/compatibility/test_openai_sdk_client_paths.py:503
↓ 7 callersFunction_chat
(client: BudgetClient, key: str, *, user: str | None = None)
tests/e2e/quota_management/budgets/test_budget_enforcement_e2e.py:32
↓ 7 callersFunction_chat_body
(model: str, marker: str, store_id: str, *, stream: bool = False)
tests/integration/providers/test_audit_any_sweep_vector_store.py:29
↓ 7 callersFunction_chat_body
(model: str, marker: str, messages: Sequence[JsonValue], stream: bool)
tests/integration/providers/test_responses_bridge_tool_search_replay_wire.py:205
↓ 7 callersFunction_chat_completion
(candidate: Gateway, model: str, key: str, marker: str)
tests/integration/observability/test_azure_storage_client_ttl.py:29
↓ 7 callersFunction_chat_completion_response
()
tests/unit/llms/github_copilot/test_per_user_auth.py:613
↓ 7 callersFunction_chat_errors
(text: str)
tests/integration/streaming/test_stream_clean_eof_without_finish_reason.py:220
↓ 7 callersFunction_chat_peer
(identity: str)
tests/integration/compatibility/test_openai_sdk_client_paths.py:138
↓ 7 callersFunction_chat_response
(text: str)
tests/unit/proxy/guardrails/guardrail_hooks/test_model_armor.py:5008
↓ 7 callersFunction_check
(tmp_path: Path, source: str)
tests/unit/test_check_unbounded_in_lists.py:21
↓ 7 callersMethod_check_file_deletion_allowed
Check if file deletion should be blocked due to batch references. Blocks deletion if: 1. File is referenced by any batch in
enterprise/litellm_enterprise/proxy/hooks/managed_files.py:2736
↓ 7 callersMethod_check_for_os_environ_vars
Check for os.environ/ variables in the config and replace them with the actual values. Includes a depth limit to prevent infinite rec
litellm/proxy/proxy_server.py:6302
↓ 7 callersFunction_check_provider_match
Check if the model info provider matches the custom provider. A missing ``litellm_provider`` key and a ``litellm_provider`` set to ``Non
litellm/utils.py:6922
↓ 7 callersMethod_check_proxy_admin_viewer_access
Check access for PROXY_ADMIN_VIEW_ONLY role. Admin Viewer follows a read-parity-with-Proxy-Admin rule: anything Proxy Admin
litellm/proxy/auth/route_checks.py:1014
↓ 7 callersFunction_check_team_model_budget_update_authority
Like `_check_team_budget_update_authority`: only a proxy admin may raise, re-window or drop a per-model cap, or add an entry that takes precedence
litellm/proxy/management_endpoints/team_endpoints.py:1529
↓ 7 callersFunction_citation
(url: str, title: str, start: int, end: int)
tests/unit/llms/azure/search/test_bing_grounding_search_transformation.py:52
↓ 7 callersFunction_claim
(job_id: str)
tests/unit/proxy/common_utils/test_check_responses_cost.py:170
↓ 7 callersFunction_claimed_post_call_pipeline_data
( *policy_names: str, extra_guardrails: dict[str, list[str]] | None = None, policy_source: str | None = "m
tests/unit/proxy/utils/proxy_logging/test_guardrail_pipeline.py:2362
↓ 7 callersFunction_classifier_config
(api_base: str, **overrides: JsonValue)
tests/integration/management/test_auto_router_databricks_classifier_writes.py:96
↓ 7 callersMethod_classifier_user_content
(mock_router_instance)
tests/unit/router_strategy/test_complexity_router.py:16766
↓ 7 callersFunction_classify_marked_test
Map cassette state → verdict tag for tests that *were* VCR-marked.
tests/_vcr_conftest_common.py:1606
↓ 7 callersFunction_cli_callback_kwargs
(flow)
tests/unit/proxy/management_endpoints/test_ui_sso.py:10522
↓ 7 callersFunction_client
(response)
tests/unit/proxy/mcp_server/outbound_credentials/test_token_endpoint.py:70
↓ 7 callersFunction_client
(app: FastAPI, root_path: str = "")
tests/unit/proxy/common_utils/test_openapi_route.py:22
↓ 7 callersFunction_client
(captured: dict[str, object], status_code: int = 200, body: dict[str, object] | None = None)
tests/unit/llms/tencent/image_generation/test_tencent_image_generation_transformation.py:20
↓ 7 callersFunction_client
(registry: CollectorRegistry)
tests/unit/integrations/test_prometheus_metrics_endpoint.py:43
↓ 7 callersFunction_client_returning
(body: str, status_code: int = 200)
tests/unit/proxy/auth/test_password_policy.py:50
↓ 7 callersFunction_client_with_transport
(handler)
tests/unit/proxy/auth/test_password_policy.py:37
↓ 7 callersFunction_clock
(ticks: tuple[float, ...])
tests/unit/router_utils/test_kubernetes_pod_discovery.py:263
↓ 7 callersFunction_coerce_response_obj_for_attrs
Return a `.get`-compatible view of `response_obj` when possible. - dicts and Pydantic models that already expose `.get` are returned unchan
litellm/integrations/arize/_utils.py:604
↓ 7 callersFunction_cohere_v2_response
(message: dict[str, object], finish_reason: str = "COMPLETE")
tests/unit/llms/cohere/chat/test_cohere_transformation.py:350
↓ 7 callersFunction_collect
(stream: AsyncIterator[bytes])
tests/unit/router_utils/test_fallback_event_handlers.py:2028
↓ 7 callersFunction_collect
(stream)
tests/unit/llms/anthropic/messages/responses_adapters/test_responses_adapters_streaming_iterator.py:477
↓ 7 callersFunction_collect_errors
()
tests/unit/test_ws_vcr.py:48
↓ 7 callersFunction_common_checks_for_customer_model
( *, model: str, customer_models: list[str], request_overrides: Mapping[str, object] | None =
tests/unit/proxy/auth/test_auth_checks_object_access_and_lookup.py:10177
↓ 7 callersFunction_complete
(redirect_uri: str, delivery, cookies=None, handle=None, session_user_id="u1")
tests/unit/proxy/mcp_server/test_gateway_dcr_flow.py:789
↓ 7 callersFunction_completed_batch_safe_to_retire
Whether a "completed" batch may be retired from cost recovery. ``batch_processed=True`` is the sole re-pickup gate for CheckBatchCost's cost-
litellm/proxy/openai_files_endpoints/common_utils.py:1630
↓ 7 callersFunction_completion
(router: Router, model: str, team: bool)
tests/integration/sdk/test_router_concurrent_deployment_delete_sdk.py:171
↓ 7 callersFunction_completion_kwargs
(provider: str, client, **overrides)
tests/unit/llms/openai/test_openai_common_utils.py:318
↓ 7 callersFunction_completion_logging_obj
(call_type: str)
tests/unit/caching/test_caching_handler.py:2699
↓ 7 callersFunction_complexity_router
( name: str = "smart", tiers: Mapping[str, object] = TIERS, deployment_id: str = "marker", **p
tests/unit/test_router/core/test_selection_characterization_prerouting.py:45
↓ 7 callersMethod_complexity_router_params
(default_model: str, tags=None)
tests/unit/test_router/test_router.py:8473
↓ 7 callersMethod_compute_protection_scopes
Call protectionScopes/compute and cache with ETag. Returns: Tuple of (etag, scope_response).
litellm/proxy/guardrails/guardrail_hooks/microsoft_purview/base.py:173
↓ 7 callersFunction_config
()
tests/unit/proxy/agent_endpoints/test_databricks_oauth.py:319
↓ 7 callersMethod_config
(self)
tests/unit/llms/github_copilot/responses/test_github_copilot_responses_transformation.py:550
↓ 7 callersFunction_config_overrides_table
(prisma_client: "PrismaClient")
litellm/proxy/management_endpoints/config_override_endpoints.py:64
↓ 7 callersFunction_configure
( proxy: Gateway, settings: Mapping[str, JsonValue], environment: Mapping[str, JsonValue] | None = None )
tests/integration/configuration/test_callback_delete.py:98
↓ 7 callersFunction_configure
( monkeypatch: pytest.MonkeyPatch, *, manager: _VaultSecrets, key_management: KeyManagementSet
tests/unit/rust_bridge/ocr/test_secrets.py:117
↓ 7 callersFunction_configured_servers
(*servers: MCPServer)
tests/unit/proxy/mcp_server/test_utils.py:27
↓ 7 callersFunction_converse_bound_thinking_request
( headers: dict[str, str] | None, body_betas: tuple[str, ...] = (), drop_params: bool = True )
tests/unit/llms/bedrock/chat/test_converse_transformation.py:12039
↓ 7 callersFunction_converse_request_for_json_schema
( model: str, stream: bool, thinking_params: Mapping[str, object] | None = None, model_info: M
tests/unit/llms/bedrock/chat/test_converse_transformation.py:4994
↓ 7 callersFunction_converse_request_for_schema
(model: str, schema: Mapping[str, object])
tests/unit/llms/bedrock/chat/test_converse_transformation.py:5510
↓ 7 callersMethod_converse_stream
(self)
tests/unit/proxy/test_common_request_processing.py:6698
↓ 7 callersFunction_converse_targets
(wire: Wire)
tests/integration/streaming/test_fake_stream_reasoning_wire.py:209
↓ 7 callersMethod_convert_raw_bytes_to_str_lines
Converts a list of raw bytes into a list of string lines, similar to aiter_lines() Args: raw_bytes: List of bytes chunks
litellm/llms/base_llm/passthrough/transformation.py:245
↓ 7 callersMethod_convert_user_api_key_auth_to_dict
Helper function to convert UserAPIKeyAuth object to dictionary. Handles both Pydantic models and regular objects.
litellm/proxy/utils.py:2243
↓ 7 callersFunction_cooled
(router: Router)
tests/unit/router_utils/test_health_check_allowed_fails_integration.py:26
↓ 7 callersFunction_count_status
(model: str = "kimi")
tests/unit/proxy/test_proxy_token_counter.py:2381
↓ 7 callersFunction_count_through_the_messages_endpoint
( router: Router, request_options: Mapping[str, JsonValue] )
tests/unit/proxy/test_proxy_token_counter.py:3734
↓ 7 callersFunction_counter
(spend_counter_store: SpendCounterStore)
tests/unit/proxy/middleware/test_budget_reservation_release_middleware.py:140
↓ 7 callersFunction_counts
(cache: DualCache)
tests/unit/test_router/test_io_token_rate_limits.py:1891
↓ 7 callersFunction_covered_models
(entries: list[JsonValue])
tests/integration/management/test_fallback_management_public_names.py:120
↓ 7 callersFunction_create_exa_tool
( gateway: Gateway, scenario: Scenario, name: str, api_base: str, input_cost_per_query: float | None = Non
tests/integration/providers/test_search_wire.py:127
↓ 7 callersMethod_create_guardrail
(self)
tests/unit/proxy/guardrails/guardrail_hooks/test_bedrock_guardrails.py:1484
↓ 7 callersFunction_create_managed_file
( router: _FakeFilesRouter, target_model_names: list[str] )
tests/unit/enterprise/proxy/test_managed_files_hook.py:2340
↓ 7 callersFunction_create_spend_logs_with_poison_isolation
Write spend-log rows, isolating any row Postgres rejects on its data. ``create_many`` writes the whole batch in a single statement, so one row
litellm/proxy/utils.py:9395
↓ 7 callersFunction_custom_fields
(row: dict[str, JsonValue])
tests/integration/management/test_model_hub_custom_metadata.py:45
↓ 7 callersMethod_custom_tier_row
(model_name: str, model_id: str)
tests/unit/router_strategy/test_complexity_router.py:1992
↓ 7 callersFunction_daily_user_spend_record
A LiteLLM_DailyUserSpend row as the per-user breakdown reads it.
tests/unit/proxy/management_endpoints/test_common_daily_activity.py:1569
↓ 7 callersFunction_db_login
A database user's sign-in with the stored hash faked, so no database or scrypt is needed.
tests/unit/proxy/auth/test_login_utils.py:727
↓ 7 callersFunction_decoded
(minted: MintedProxyCredential)
tests/unit/proxy/mcp_server/test_proxy_api_credentials.py:43
↓ 7 callersFunction_decoded
(model_id: str, deployment_id: str | None)
tests/unit/proxy/video_endpoints/test_utils.py:467
↓ 7 callersFunction_decoded_frames
(raw: bytes)
tests/unit/llms/bedrock/passthrough/guardrail_translation/test_handler.py:2531
↓ 7 callersFunction_deduplicate_bedrock_content_blocks
Remove duplicate content blocks that share the same ID under ``block_key``. Bedrock requires all toolResult and toolUse IDs within a single
litellm/litellm_core_utils/prompt_templates/factory.py:4583
↓ 7 callersFunction_default_of
(func, param)
tests/unit/proxy/client/cli/test_agents.py:53
↓ 7 callersFunction_delete_callback_roundtrip
POST /config/callback/delete as admin against a stubbed config, and return (response, saved litellm_settings) so each case pins the full body
tests/unit/proxy/proxy_server/test_routes_config.py:1002
↓ 7 callersFunction_delivered
(wire: Wire)
tests/integration/providers/test_vertex_anthropic_structured_output_wire.py:604
↓ 7 callersFunction_denied_scoped_list
( user_api_key_auth: UserAPIKeyAuth, mcp_servers: list[str], mock_manager: MagicMock, resolver
tests/unit/proxy/mcp_server/test_mcp_server_tool_calls_and_headers.py:1573
↓ 7 callersFunction_deployment
(gateway: Gateway, scenario: Scenario, peer: BedrockRealtimePeer)
tests/integration/providers/test_bedrock_nova_sonic_realtime_tool_calls.py:133
↓ 7 callersFunction_deployment
(scenario: Scenario, provider: Provider, wire_url: str)
tests/integration/providers/test_responses_bridge_tool_search_replay_wire.py:133
↓ 7 callersFunction_deployment
(info: Mapping[str, object], identity: str = "embedding")
tests/unit/router_utils/test_endpoint_eligibility.py:17
↓ 7 callersFunction_deployment
(windows: object = None)
tests/unit/router_utils/test_access_windows.py:29
↓ 7 callersFunction_deployment
(model_name: str, model: str = "openai/gpt-4o", **model_info)
tests/unit/proxy/test_model_list_callback_filter.py:65
↓ 7 callersFunction_diff
(path: str, *hunks: str)
tests/unit/test_spend_logs_guard.py:44
↓ 7 callersMethod_discard_deferred_stream_logging_for_failure
Drop the parked success dispatch for an assembled chat stream or a stream that billed several rounds when it ends in an error ``post_call_fail
litellm/proxy/utils.py:5274
↓ 7 callersFunction_dispatch_audit_log_to_callbacks
Dispatch audit log to all registered audit_log_callbacks.
litellm/proxy/management_helpers/audit_logs.py:188
↓ 7 callersFunction_dispatched_failure_logging_obj
A Logging object as the failure hook sees it: litellm_params in the real get_litellm_params() shape (every key present, ``litellm_metadata`` None
tests/unit/proxy/test_proxy_utils_model_creation_and_error_logging.py:3562
↓ 7 callersFunction_done
(value: T)
tests/unit/redis/stores/spend_counters_double.py:17
↓ 7 callersFunction_drain
()
tests/unit/llms/custom_httpx/test_aiohttp_transport.py:136
↓ 7 callersFunction_drain_logging_worker
Run every queued logging task to completion on the current event loop. The success event is delivered through the fire-and-forget GLOBAL_LOGGING_
tests/unit/rag/test_main.py:36
↓ 7 callersFunction_drain_sse
(wrapper: AnthropicStreamWrapper)
tests/unit/llms/anthropic/messages/adapters/test_streaming_iterator_mid_stream_error.py:81
↓ 7 callersFunction_drain_sync
(wrapper: CustomStreamWrapper)
tests/unit/litellm_core_utils/test_streaming_overhead.py:104
↓ 7 callersFunction_drive_to_block
Spend until the cap blocks a call, staying under one window so the block is observed before the reset job can fire; fail hard if enforcement never
tests/e2e/quota_management/budgets/test_budget_reset_e2e.py:26
← previousnext →9,901–10,000 of 202,479, ranked by callers