MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 6 callersFunction_set_salt_key
(monkeypatch)
tests/unit/proxy/common_utils/test_callback_utils.py:370
↓ 6 callersMethod_set_spend_logs_row_count_in_proxy_state
Set the `LiteLLM_SpendLogs`row count in proxy state. This is used later to determine if we should run expensive UI Usage queries.
litellm/proxy/utils.py:8227
↓ 6 callersFunction_set_usage_outputs
(span: "Span", response_obj, span_attrs)
litellm/integrations/arize/_utils.py:260
↓ 6 callersFunction_set_user_keys_blocked
Block or unblock virtual keys owned by a user and invalidate them in the in-memory/redis caches so the change takes effect immediately.
litellm/proxy/management_endpoints/scim/scim_v2.py:1192
↓ 6 callersFunction_settled
(terminal: int, posts: int)
tests/unit/test_router/test_router_callback_hook_sequence.py:240
↓ 6 callersFunction_settled_stats
(database_url: str, seeded_rows: int | None = None)
tests/integration/spend/test_key_metadata_recovery_probe_bounds.py:83
↓ 6 callersFunction_settlement
(interaction_id: str)
tests/integration/spend/test_background_interaction_settlement.py:304
↓ 6 callersFunction_setup_block_unblock_mocks
Helper to set up common mocks for block/unblock tests.
tests/unit/proxy/management_endpoints/test_key_management_endpoints.py:11584
↓ 6 callersFunction_setup_db
Create a database instance with a stubbed prisma client.
tests/unit/integrations/focus/test_focus_database.py:13
↓ 6 callersFunction_setup_proxy_logging
Patch proxy_logging_obj so failure hook can be asserted.
tests/unit/responses/mcp/test_litellm_proxy_mcp_handler.py:72
↓ 6 callersFunction_shared_ptu_router
(model_group: str)
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:8903
↓ 6 callersFunction_shell
()
tests/integration/authorization/test_tool_search_replay_allowlist.py:20
↓ 6 callersFunction_shipped_token_count_model_field
(model: str, field: str)
tests/integration/providers/test_count_tokens_model_wire.py:59
↓ 6 callersMethod_should_block_content
Check if Model Armor response indicates content should be blocked, including both inspectResult and deidentifyResult.
litellm/proxy/guardrails/guardrail_hooks/model_armor/model_armor.py:440
↓ 6 callersMethod_should_disable_vad_auto_response
Transcription and translation sessions have no assistant turn to gate and reject realtime session updates.
litellm/litellm_core_utils/realtime_streaming.py:1640
↓ 6 callersFunction_should_track_cost_callback
Determine if the cost callback should be tracked based on the kwargs Pass-through endpoints can be configured with ``auth=false``, which lea
litellm/proxy/hooks/proxy_track_cost_callback.py:1144
↓ 6 callersFunction_sign
( config: StrandsDeciderDecisionsConfig, api_base: str, headers: Mapping[str, str], api_key: s
tests/unit/llms/strands_decider/decisions/test_strands_decider_decisions_transformation.py:41
↓ 6 callersFunction_sign_claims
(payload: dict, prefix: str = SESSION_TOKEN_PREFIX, keys: SessionKeys = KEYS)
tests/unit/proxy/mcp_server/outbound_credentials/test_session_token.py:76
↓ 6 callersFunction_signature_deltas
(events: List[dict])
tests/unit/llms/anthropic/messages/adapters/test_streaming_iterator_first_delta.py:311
↓ 6 callersFunction_signed_batch_request
(config, litellm_params: dict, optional_params: dict)
tests/unit/llms/bedrock/batches/test_transformation.py:276
↓ 6 callersFunction_silence_signal_handling
(monkeypatch)
tests/unit/proxy/client/cli/autoroute/test_commands.py:44
↓ 6 callersFunction_single
(gateway: Gateway, tool_name: str)
tests/integration/management/test_tool_policy_user.py:225
↓ 6 callersFunction_sink
(exporter: Exporter, clock: _Clock, max_rows: int = 100)
tests/unit/telemetry/test_aggregate.py:71
↓ 6 callersFunction_skill
(skill_id: str, created_by: str | None)
tests/unit/llms/litellm_proxy/test_skills_ownership.py:29
↓ 6 callersFunction_skipping_system
(guardrail: CustomGuardrail)
tests/unit/llms/openai/responses/test_openai_responses_guardrail_handler.py:508
↓ 6 callersFunction_smart_router_model_list
()
tests/unit/router_strategy/test_router_routing_plugins.py:71
↓ 6 callersFunction_span
(name: str, *, failed: bool = False, ttft: float | None = None)
tests/e2e_harness/logging/test_span_selection.py:30
↓ 6 callersMethod_span
(self)
tests/unit/integrations/test_opentelemetry.py:5871
↓ 6 callersFunction_span_id
()
tests/integration/observability/test_otel_trace_response_header.py:56
↓ 6 callersMethod_spawn_engine_task
(self, work: Coroutine[object, object, object])
litellm/proxy/utils.py:7086
↓ 6 callersFunction_spec_with
`replace` re-runs `__init__`, so the spec's own validation applies to the override.
tests/unit/proxy/list_api/test_list_framework.py:84
↓ 6 callersFunction_spend_capture_logger
()
tests/unit/proxy/common_utils/test_check_responses_cost.py:1698
↓ 6 callersFunction_spend_log_for_call_type
( call_type: str, internal_call_origin: str | None = None, background: bool | None = None, man
tests/unit/proxy/spend_tracking/test_spend_tracking_utils.py:5192
↓ 6 callersFunction_spend_log_reads
The raw reads ``/spend/logs/ui`` ran under the statement timeout.
tests/unit/proxy/spend_tracking/test_spend_query_optimization.py:206
↓ 6 callersFunction_spend_log_reads
The raw reads the cache dashboard ran under the statement timeout.
tests/unit/proxy/analytics_endpoints/test_analytics_endpoints.py:78
↓ 6 callersFunction_spend_row
(identity: str)
tests/integration/providers/test_custom_prompt_template_wire.py:125
↓ 6 callersFunction_spend_row
(*request_ids: str)
tests/integration/providers/test_gemini_tool_result_part_reference_wire.py:525
↓ 6 callersFunction_spend_row
(key: str)
tests/integration/spend/test_passthrough_upstream_model.py:56
↓ 6 callersFunction_spend_rows_matching_call
( rows: tuple[dict[str, JsonValue], ...], model: str, call_id: str, )
tests/integration/observability/_logging_only_scope_support.py:1079
↓ 6 callersMethod_split_text_for_analysis
Split ``text`` into chunks whose JSON-serialized form is at most ``chunk_size_bytes`` bytes (the analyzer body limit applies to the
litellm/proxy/guardrails/guardrail_hooks/presidio.py:721
↓ 6 callersFunction_sse
(events: Sequence[Mapping[str, object]])
tests/unit/responses/mcp/test_aresponses_api_with_mcp.py:1473
↓ 6 callersFunction_sse_event
(event_type: str, data: dict)
tests/unit/llms/anthropic/messages/native/test_agentic_streaming_iterator.py:35
↓ 6 callersFunction_sse_payloads
(collected: Tuple[StreamChunk, ...])
tests/unit/proxy/guardrails/guardrail_hooks/unified_guardrails/test_openai_streaming_block.py:176
↓ 6 callersFunction_sse_reply
(body: str)
tests/unit/llms/openai/responses/test_openai_responses_http.py:174
↓ 6 callersFunction_start_body
(media_uri: str = "s3://tenant-media/call.wav", **members: object)
tests/unit/proxy/pass_through_endpoints/llm_provider_handlers/test_transcribe_passthrough_logging_handler.py:531
↓ 6 callersFunction_start_cli_sso_flow
(base_url: str)
litellm/proxy/client/cli/commands/auth.py:625
↓ 6 callersMethod_start_partial_stream
(self)
tests/unit/proxy/test_common_request_processing.py:9090
↓ 6 callersFunction_state
(marker: str, scenario: str = "plain")
tests/integration/providers/test_cloudflare_clairic_extraction_wire.py:42
↓ 6 callersFunction_status_code_for_error_fields
(error_type: str | None, error_code: str | None)
litellm/responses/streaming_iterator.py:305
↓ 6 callersFunction_stock_function_tool
()
tests/integration/providers/test_openai_responses_tool_search_wire.py:71
↓ 6 callersFunction_storage
(url: str, retention_days: int = 14)
tests/test_litellm_rust/test_clickhouse_spend.py:17
↓ 6 callersFunction_stored_response_turn
( request_id: str, call_type: str, status: str, output: list[dict[str, object]], prompt: str | None = None
tests/unit/responses/litellm_completion_transformation/test_session_handler.py:1086
↓ 6 callersFunction_stored_settings
(database_url: str)
tests/integration/configuration/test_callback_delete.py:87
↓ 6 callersFunction_str_keyed
(value: object)
litellm/router_utils/pre_call_checks/continuation_prefill_check.py:37
↓ 6 callersFunction_strategy_source
( *, strategy_id: str = "example", surfaces: tuple[str, ...] = (), drop: tuple[str | None, str
tests/rust-python-harness/cli/test_cli.py:29
↓ 6 callersFunction_stream
(rig: Rig, guardrail: str, marker: str, model: str = CLAUDE)
tests/integration/observability/test_bedrock_invoke_stream_post_call_guardrail.py:470
↓ 6 callersFunction_stream_builder_logging_obj
(model: str = "gpt-4o", custom_llm_provider: str = "openai")
tests/unit/test_main.py:5895
↓ 6 callersFunction_stream_chat
(gateway: Gateway, model: str, key: str)
tests/integration/streaming/test_openai_like_stream_usage_wire.py:151
↓ 6 callersFunction_streamable_post_call_pipelines
The post_call pipelines a streaming response can be gated through. Streaming pipelines scan the buffered stream through the endpoint guardra
litellm/proxy/utils.py:1156
↓ 6 callersFunction_streamed_text
(events: tuple[Any, ...])
tests/unit/llms/base_llm/responses/test_transformation.py:803
↓ 6 callersFunction_streaming_logging_obj_with_callbacks
(callbacks: list[CustomLogger])
tests/unit/litellm_core_utils/test_litellm_logging.py:8828
↓ 6 callersFunction_strip_routing_prefix
(tags: Sequence[str], prefix: str)
litellm/router_strategy/tag_based_routing.py:203
↓ 6 callersFunction_strip_total_tokens_from_anthropic_response
Remove the OpenAI-flavored `usage.total_tokens` field that LiteLLM injects into Anthropic /v1/messages responses. The Anthropic /v1/messages
litellm/proxy/anthropic_endpoints/endpoints.py:96
↓ 6 callersFunction_stub_handler
Build a stub AgentsHTTPHandler whose CRUD methods return *return_value*.
tests/unit/interactions/test_agents_main_and_utils.py:116
↓ 6 callersFunction_styled_input
Like input() but wraps ANSI sequences in readline ignore markers (\\001...\\002) so readline correctly tracks the cursor column. In non-T
litellm/setup_wizard.py:177
↓ 6 callersFunction_success_kwargs
(preset_cache_key: str | None = "preset-key")
tests/unit/proxy/spend_tracking/test_spend_event.py:51
↓ 6 callersFunction_support_agent_router
(model: str = "azure_ai/agents/support-agent")
tests/unit/llms/azure_ai/responses/test_azure_ai_responses_transformation.py:427
↓ 6 callersMethod_supports_reasoning_effort_level
Check if the model supports a specific reasoning_effort level. Looks up ``supports_{level}_reasoning_effort`` in the model map via th
litellm/llms/openai/chat/gpt_reasoning_transformation.py:187
↓ 6 callersFunction_surface_prompts
(marker: str, per_surface: int)
tests/integration/observability/test_s3_v2_partition_granularity.py:171
↓ 6 callersFunction_sync
A member both cores serve outside the ``RouterCore`` protocol (an endpoint or a deprecated name).
tests/unit/test_router/core/test_parity_rulings_2026_11.py:41
↓ 6 callersFunction_sync_chat
(rig: Rig, text: str)
tests/integration/observability/test_content_filter_logging_only.py:268
↓ 6 callersFunction_sync_client
(captured_requests: list[httpx.Request], response: httpx.Response)
tests/unit/llms/together_ai/chat/test_together_ai_chat_transformation.py:810
↓ 6 callersMethod_sync_client
( self, litellm_params: GenericLiteLLMParams, client: HTTPHandler | None, )
litellm/interactions/http_handler.py:60
↓ 6 callersFunction_tag_router
(model_list: list[dict[str, object]], tag_routing_prefix: str = "route:")
tests/unit/router_strategy/test_router_tag_routing.py:3436
↓ 6 callersFunction_tag_url
(tag: str)
tests/unit/proxy/ui_crud_endpoints/test_latest_release_endpoints.py:79
↓ 6 callersFunction_team_membership_db
(prisma_client: PrismaClient | None)
litellm/proxy/management_endpoints/team_endpoints.py:544
↓ 6 callersFunction_team_model_row
(public_name: str, litellm_params: Mapping[str, object])
tests/unit/proxy/management_endpoints/test_model_management_endpoints.py:8266
↓ 6 callersFunction_team_requests_flushed
(team: str, count: int)
tests/integration/spend/test_disable_end_user_cost_tracking.py:367
↓ 6 callersFunction_team_roster_events
(audit_logger: _RecordingAuditLogger, action: str)
tests/unit/proxy/management_endpoints/test_team_endpoints.py:15102
↓ 6 callersFunction_team_rows
(team_id: str)
tests/integration/management/test_ptu_share_guards.py:98
↓ 6 callersFunction_tenants
(scenario: Scenario, purpose: str)
tests/integration/management/test_managed_file_create_ownership.py:96
↓ 6 callersFunction_test_guardrail_definitions
Instantiate and run each guardrail definition against the text.
litellm/proxy/management_endpoints/policy_endpoints/endpoints.py:1219
↓ 6 callersFunction_text
(value: object)
litellm/accounting/projection.py:48
↓ 6 callersFunction_text_completion_stream
(upstream: Sequence[Completion])
tests/unit/text_completions/test_streaming_iterator.py:102
↓ 6 callersFunction_text_delta
(text: str, sequence_number: int)
tests/unit/responses/test_responses_streaming_iterator.py:441
↓ 6 callersFunction_text_parts
(request: Request)
tests/integration/providers/test_realtime_webrtc_http.py:125
↓ 6 callersFunction_thinking_block
(signature: JsonValue, thinking: str = _REASONING)
tests/integration/providers/test_gemini_thinking_replay_wire.py:69
↓ 6 callersFunction_thinking_param
(case: _Case, body: Mapping[str, JsonValue])
tests/unit/llms/base_llm/chat/test_provider_chat_thinking.py:221
↓ 6 callersMethod_to_model
Convert a database record to a Team model.
litellm/repositories/team_repository.py:118
↓ 6 callersFunction_token_hash_for_update
Resolve the token hash to store, or None to leave the mapped key alone.
litellm/proxy/management_endpoints/jwt_key_mapping_endpoints.py:77
↓ 6 callersFunction_tool_call
(tool_call_id, name)
tests/unit/llms/bedrock/chat/chat_completions/test_bedrock_chat_completions_transformation.py:1633
↓ 6 callersFunction_tool_call_request
(model: str, tool_name: str)
tests/integration/management/test_tool_policy_user.py:32
↓ 6 callersFunction_tool_list
(*tools: dict[str, JsonValue])
tests/unit/responses/test_additional_tools.py:60
↓ 6 callersFunction_tool_message
(tool_call_id: JsonValue, content: str)
tests/integration/providers/test_responses_bridge_tool_search_replay_wire.py:172
↓ 6 callersFunction_tool_outputs_kept
(view: InferenceRequestView)
tests/unit/llms/anthropic/messages/context_management/test_inference_view.py:123
↓ 6 callersFunction_tool_policy_config
(directory: Path, model: str, upstream_url: str, mode: str | list[str] = "pre_call")
tests/integration/management/test_tool_policy_user.py:519
↓ 6 callersFunction_tools
(*tools: JsonValue)
tests/integration/sdk/test_responses_bridge_tool_search_sdk.py:23
↓ 6 callersFunction_toolset_from_row
(row: MCPToolsetRow)
litellm/proxy/mcp_server/toolset_db.py:73
↓ 6 callersFunction_toolset_table
The toolset table actions of the prisma client.
litellm/proxy/mcp_server/toolset_db.py:68
↓ 6 callersMethod_toolset_tools_for_server
Tool names this row's toolsets grant on ``server_id``, ``None`` when its toolsets place no restriction on that server (it declares no toolsets
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:3001
← previousnext →12,401–12,500 of 202,479, ranked by callers