MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 5 callersFunction_enqueued_test_handler
()
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:7533
↓ 5 callersMethod_ensure_output_item_for_chunk
(self, chunk: ModelResponseStream)
litellm/responses/litellm_completion_transformation/streaming_iterator.py:1205
↓ 5 callersFunction_entra_error
(aadsts_code: str, description: str)
tests/unit/llms/azure/test_azure_common_utils.py:2727
↓ 5 callersFunction_entries
(response: httpx.Response)
tests/integration/configuration/test_codex_catalog_boot.py:62
↓ 5 callersFunction_entry
(capabilities: CodexCapabilities, **kwargs: object)
tests/unit/proxy/common_utils/test_codex_model_translation.py:21
↓ 5 callersFunction_env
(repo: Path, bin_dir: Path, extra_env: dict[str, str])
tests/unit/test_pre_commit_lint.py:134
↓ 5 callersFunction_environment_variables_proxy_config
( directory: Path, gateway: Gateway, alias: str, config_file_provider: Wire )
tests/integration/configuration/test_effective_settings.py:1154
↓ 5 callersFunction_environment_variables_proxy_environment
(database_url: str, process_key: str)
tests/integration/configuration/test_effective_settings.py:1170
↓ 5 callersFunction_error_body
(message: str)
tests/integration/providers/test_bedrock_converse_stream_event_frames_wire.py:70
↓ 5 callersFunction_error_message
(exc: BaseException)
litellm/proxy/management_helpers/bulk_user_creation.py:189
↓ 5 callersFunction_error_message
(text: str)
tests/integration/providers/test_ollama_prompt_tools_wire.py:941
↓ 5 callersMethod_error_messages
(caplog: pytest.LogCaptureFixture)
tests/unit/litellm_proxy_extras/test_litellm_proxy_extras_utils.py:1970
↓ 5 callersFunction_error_summary
(error: ExchangeError)
litellm/llms/base_llm/auth/token_exchange.py:232
↓ 5 callersFunction_estimate_dispatched_failure_usage
A request that failed after dispatch consumed provider-billed input tokens, but no provider usage ever came back. Estimate the input side with
litellm/proxy/utils.py:1223
↓ 5 callersFunction_estimate_key_row
(token: str, metadata: dict)
tests/unit/proxy/management_endpoints/test_key_management_endpoints.py:19484
↓ 5 callersMethod_event_stream_translation
The provider handler for a relayed event stream: the one that owns its frames, else ``provider``'s.
litellm/llms/pass_through/guardrail_translation/handler.py:314
↓ 5 callersFunction_every_carrier
(model: str, name: str, guardrail: str, stream: bool)
tests/integration/messages_endpoint/test_guardrail_long_tool_names_wire.py:45
↓ 5 callersFunction_everything_on_disk
(home)
tests/unit/litellm_core_utils/test_cli_token_utils.py:59
↓ 5 callersFunction_exchange_for_bridge_server_with_jwt
Drive exchange_token_with_server for a bridge oauth_delegate authorization_code request whose presented credential is JWT-shaped, with _resolve_jw
tests/unit/proxy/mcp_server/test_discoverable_endpoints.py:7587
↓ 5 callersMethod_exec_capture
(self, cmd: Sequence[str], *, input: bytes | None = None)
litellm/harness/sandbox/docker.py:184
↓ 5 callersMethod_execute_delete_batch
Run one delete batch under a Postgres statement and lock timeout. The timeouts are what actually bound the work: a Prisma transactio
litellm/proxy/db/db_transaction_queue/spend_log_cleanup.py:757
↓ 5 callersFunction_expect_execution_timeout
(guardrail: CustomCodeGuardrail)
tests/unit/proxy/guardrails/test_custom_code_security.py:644
↓ 5 callersFunction_expected
(mode: Mode, marker: str, *forwarded: JsonValue)
tests/integration/providers/test_responses_bridge_input_audio_wire.py:126
↓ 5 callersFunction_explicitly_free
(name: str, model: str = "gpt-3.5-turbo", **model_info: float)
tests/unit/proxy/auth/test_unmapped_model_budget_enforcement.py:22
↓ 5 callersFunction_extract
( gateway: Gateway, model: str, marker: str, *, route: Route = "/v1/extraction", key:
tests/integration/providers/test_cloudflare_clairic_extraction_wire.py:99
↓ 5 callersMethod_extract_anonymized_content
Extract anonymized content from Noma API response. Args: response_json: The full response from Noma API mess
litellm/proxy/guardrails/guardrail_hooks/noma/noma.py:409
↓ 5 callersMethod_extract_mcp_server_names_from_tools
Extract MCP server names from tools with type=mcp and litellm_proxy server_url.
litellm/proxy/guardrails/guardrail_hooks/mcp_security/mcp_security_guardrail.py:79
↓ 5 callersFunction_extract_response_text
Extract text from every LLM response choice.
litellm/proxy/guardrails/guardrail_hooks/semantic_guard/semantic_guard.py:197
↓ 5 callersMethod_extract_status_code
Extract HTTP status code from various input formats for validation. This is a centralized helper to extract status code from differe
litellm/integrations/prometheus.py:2809
↓ 5 callersFunction_extract_summary_text
(raw: str | None)
litellm/llms/anthropic/messages/context_management/editors/compact.py:403
↓ 5 callersMethod_extract_tags
Builds the list of tags for a Datadog metric point
litellm/integrations/datadog/datadog_metrics.py:61
↓ 5 callersMethod_extract_text_for_guardrail
Extract text from data for guardrail processing. If field_expressions provided, extracts only those fields. Otherwise, retur
litellm/llms/pass_through/guardrail_translation/handler.py:61
↓ 5 callersMethod_extract_thinking_blocks_from_parts
Extract thinking blocks from parts if present. Per Google's docs (https://ai.google.dev/gemini-api/docs/thinking): - Parts with `thou
litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py:2145
↓ 5 callersMethod_extract_thought_signatures_from_parts
Extract thoughtSignature values from parts. Per Google's docs, thoughtSignature is returned for multi-turn context preservation and c
litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py:2168
↓ 5 callersMethod_extract_user_api_key_metadata
Extract user API key metadata from request_data for the DeepKeep API. Args: request_data: Request data dictionary contai
litellm/proxy/guardrails/guardrail_hooks/deepkeep/deepkeep.py:183
↓ 5 callersFunction_extract_user_text
Extract the latest user message text.
litellm/proxy/guardrails/guardrail_hooks/semantic_guard/semantic_guard.py:180
↓ 5 callersFunction_facet_problem
(query: str, spec: ListSpec[BudgetRow, BudgetOut] | None = None)
tests/unit/proxy/list_api/test_list_framework.py:916
↓ 5 callersFunction_failed_call_row
( rig: _StoredRig, path: str, marker: str, headers: Mapping[str, str], key: str | None = N
tests/integration/providers/test_responses_compaction_items_wire.py:981
↓ 5 callersFunction_failure_message
(call_id: str)
tests/integration/providers/test_bedrock_mantle_context_overflow_wire.py:246
↓ 5 callersFunction_failure_payload_after_mid_stream_timeout
( monkeypatch: pytest.MonkeyPatch, request_input: str | list[dict[str, object]], upstream: _Upstre
tests/unit/responses/test_responses_streaming_iterator.py:275
↓ 5 callersFunction_fake_request
(cookies: dict[str, str] | None = None)
tests/unit/proxy/management_endpoints/test_saml_sso.py:180
↓ 5 callersFunction_fallback_model_names
Model names the auth check validates for a top-level ``fallbacks`` value.
tests/unit/proxy/auth/test_router_override_fallback_auth.py:18
↓ 5 callersMethod_fan_out_of
(logger: OpenTelemetryV2)
tests/unit/integrations/otel/test_otel_v2_destinations.py:1962
↓ 5 callersFunction_fetch_with_single_outcome
(monkeypatch, outcome)
tests/unit/litellm_core_utils/test_get_model_cost_map.py:1010
↓ 5 callersFunction_field_of
(value: object, name: str)
litellm/litellm_core_utils/streaming_chunk_builder_utils.py:1507
↓ 5 callersFunction_field_reader
(obj: object)
litellm/integrations/shadow_eval_logger.py:186
↓ 5 callersFunction_fields
(value: object)
litellm/accounting/projection.py:42
↓ 5 callersFunction_file
(**overrides)
tests/unit/llms/mistral/files/test_mistral_files_transformation.py:24
↓ 5 callersFunction_file_table
(prisma_client: PrismaClient)
litellm/proxy/pass_through_endpoints/managed_id_rewriter.py:298
↓ 5 callersFunction_files_path
(provider: str | None)
tests/e2e/batches/batch_client.py:268
↓ 5 callersMethod_fill_missing_fields
Heuristically fill missing required fields with safe defaults based on the event model's field annotations. This keeps parsing tolera
litellm/llms/volcengine/responses/transformation.py:446
↓ 5 callersMethod_filter_headers_for_aws_signature
Filter headers to only include those that AWS SigV4 includes in signature calculation. This Fixes forwarded client headers from break
litellm/llms/bedrock/base_aws_llm.py:1613
↓ 5 callersMethod_finalize_refused_agentic_response
Turn the response into a terminal turn after a safety rail refused the rerun. The refused tool calls target tools LiteLLM injected o
litellm/llms/custom_httpx/llm_http_handler.py:6091
↓ 5 callersFunction_find_expires_clauses
Recursively collect every value keyed 'expires' anywhere in a Prisma where dict.
tests/unit/proxy/management_endpoints/test_key_management_endpoints.py:18884
↓ 5 callersMethod_find_jobs
(self, where: Mapping[str, object])
enterprise/litellm_enterprise/proxy/common_utils/check_batch_cost.py:179
↓ 5 callersFunction_find_key_anywhere
(obj, key: str)
tests/unit/llms/azure_ai/chat/test_azure_ai_transformation.py:822
↓ 5 callersFunction_fire
(manager: MCPServerManager, server: MCPServer, n: int)
tests/unit/proxy/mcp_server/test_mcp_max_concurrent_requests.py:137
↓ 5 callersFunction_fireworks_responses_route
(model: str)
tests/unit/llms/fireworks_ai/responses/test_fireworks_ai_responses_transformation.py:85
↓ 5 callersFunction_fixture
()
tests/unit/secret_managers/test_cyberark_secret_manager.py:38
↓ 5 callersFunction_flag_model
(monkeypatch: pytest.MonkeyPatch, name: str, **flags: bool)
tests/unit/rust_bridge/host/test_model_capabilities.py:11
↓ 5 callersFunction_flags
(generated: Mapping[str, JsonValue], slug: str)
tests/code_coverage_tests/test_generate_provider_endpoints_support.py:41
↓ 5 callersFunction_flow_handle_for
(user_id="user-a", credential_name="copilot-cred", expires_at=None)
tests/unit/proxy/credential_endpoints/test_endpoints.py:2830
↓ 5 callersMethod_flush_daily_spend_queue
( self, queue: DailySpendUpdateQueue, entity_type: Literal["user", "team", "org", "tag
litellm/proxy/db/db_spend_update_writer.py:1984
↓ 5 callersMethod_flush_pending_messages_until_setup
(self)
litellm/litellm_core_utils/realtime_streaming.py:1445
↓ 5 callersFunction_for_project
(observed: tuple[dict[str, JsonValue], ...], project: str)
tests/integration/providers/test_vertex_realtime_multi_region_host.py:227
↓ 5 callersFunction_form_xobject
(resources_number: int, drawing: bytes = _DRAW_X1)
tests/unit/litellm_core_utils/test_token_counter.py:2793
↓ 5 callersFunction_format_byok_openapi_auth_header
Format a raw BYOK credential for OpenAPI tool ``Authorization`` injection. A non-BYOK server short-circuits ``_resolve_byok_mcp_auth_header``, so
litellm/proxy/mcp_server/mcp_server_manager.py:1377
↓ 5 callersMethod_forward
(self)
tests/rust-python-harness/shared/parity/fixtures/recording.py:101
↓ 5 callersMethod_forward_bedrock_to_client
Forward messages from Bedrock to the client, yielding the ones to record for spend logging. Provider failures propagate to the caller; the cl
litellm/llms/bedrock/realtime/handler.py:901
↓ 5 callersFunction_forward_environment
(forward_url: str, **extra: str)
tests/integration/providers/test_voyage_mongodb_host_wire.py:133
↓ 5 callersFunction_forwarding
(key: str)
tests/integration/authorization/test_rag_team_provider_credentials.py:2051
↓ 5 callersFunction_frame
( identity: str, choices: list[Mapping[str, object]], usage: Mapping[str, JsonValue] | None = None
tests/integration/providers/test_databricks_chat_wire.py:166
↓ 5 callersFunction_frame
(identity: str, choices: list[JsonValue], usage: dict[str, JsonValue] | None = None)
tests/integration/messages_endpoint/chat_bridge/test_chat_bridge_multi_tool_call_blocks_wire.py:180
↓ 5 callersFunction_frame
(payload: dict)
tests/unit/proxy/pass_through_endpoints/test_injected_stream_usage.py:15
↓ 5 callersFunction_fruit_word_filter
()
tests/unit/proxy/guardrails/guardrail_hooks/content_filter/test_content_filter.py:3677
↓ 5 callersFunction_gateway_authenticator
()
tests/unit/llms/chatgpt/responses/test_chatgpt_responses_transformation.py:441
↓ 5 callersFunction_gecko_prediction
(values: object, token_count: int)
tests/unit/llms/vertex_ai/vertex_embeddings/test_embedding_handler.py:66
↓ 5 callersFunction_gemini_deployment
(model_name: str, litellm_params: Mapping[str, JsonValue])
tests/unit/proxy/test_proxy_token_counter.py:3727
↓ 5 callersFunction_gemini_frames
(response_text: str)
tests/integration/providers/test_google_genai_adapter_wire.py:818
↓ 5 callersFunction_generate_bodies
(wire: Wire)
tests/integration/providers/test_custom_prompt_template_wire.py:98
↓ 5 callersFunction_generate_content_body
()
tests/unit/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py:7811
↓ 5 callersFunction_generate_content_payload
()
tests/unit/proxy/pass_through_endpoints/llm_provider_handlers/test_vertex_passthrough_logging_handler.py:132
↓ 5 callersMethod_generate_file_name
Generate file name based on naming strategy.
litellm/llms/base_llm/files/azure_blob_storage_backend.py:101
↓ 5 callersFunction_generate_key
A key the team delete is expected to remove; cleanup only deletes it if it is still live.
tests/integration/management/test_team_delete_large_membership.py:210
↓ 5 callersFunction_generate_response
( client: ManagementClient, resources: ResourceManager, body: KeyGenerateBody )
tests/e2e/management/test_management_e2e.py:1083
↓ 5 callersMethod_generate_stable_server_id
Generate a stable server ID based on server parameters using a hash function. This is critical to ensure the server_id is stable acr
litellm/proxy/mcp_server/mcp_server_manager.py:7562
↓ 5 callersFunction_generated
(stream: bool)
tests/unit/llms/vertex_ai/context_caching/test_known_cached_contents.py:427
↓ 5 callersMethod_get_actor
Return a stable actor identifier from request metadata (e.g. user email or id).
litellm/proxy/guardrails/guardrail_hooks/ovalix/ovalix.py:204
↓ 5 callersMethod_get_agent_object_permission
Get agent object_permission via the established ``get_object_permission`` helper. Caches the ``agent_id -> object_permission_id`` map
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:4493
↓ 5 callersMethod_get_allowed_mcp_servers_for_user
The MCP servers the internal user is entitled to, as server ids. ``[]`` means this human places no restriction (allow-all from this level); `
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:4191
↓ 5 callersMethod_get_application_id
( guardrail: NomaGuardrail, user_auth: UserAPIKeyAuth, request_data: dict, ext
tests/unit/proxy/guardrails/guardrail_hooks/test_noma.py:220
↓ 5 callersMethod_get_batch_rate_limiter
Get or lazy-load the batch rate limiter.
litellm/proxy/hooks/parallel_request_limiter_v3.py:766
↓ 5 callersFunction_get_bearer_token_or_received_api_key
(api_key: str)
litellm/proxy/auth/user_api_key_auth.py:1127
↓ 5 callersMethod_get_budget_config_for_deployment
( self, model_id: object, )
litellm/router_strategy/budget_limiter.py:796
↓ 5 callersMethod_get_budget_config_for_provider
(self, provider: str)
litellm/router_strategy/budget_limiter.py:804
↓ 5 callersFunction_get_customer_id_from_standard_headers
Check standard customer ID headers for a customer/end-user ID. This enables tools like Claude Code to pass customer IDs via ANTHROPIC_CUSTOM
litellm/proxy/auth/auth_utils.py:1934
↓ 5 callersMethod_get_effective_flags
Return (enable_user_key_affinity, enable_responses_api_affinity, enable_session_id_affinity) for the given model group. If t
litellm/router_utils/pre_call_checks/deployment_affinity_check.py:124
↓ 5 callersFunction_get_endpoint_exception_status
Return the HTTP status code for an unhealthy endpoint. Prefers the live exception object in `exceptions` (direct health check path). Falls ba
litellm/proxy/proxy_server.py:4828
↓ 5 callersMethod_get_enterprise_route_checks
Import EnterpriseRouteChecks from the local enterprise source file.
tests/unit/proxy/auth/test_route_checks.py:2743
↓ 5 callersFunction_get_fields_from_model
Get the fields from a Pydantic model as a nested dictionary structure
litellm/proxy/guardrails/guardrail_endpoints.py:2018
← previousnext →14,301–14,400 of 202,479, ranked by callers