MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 6 callersFunctionspend_storage_config
(environ: Mapping[str, str] = os.environ)
litellm/rust_bridge/clickhouse.py:17
↓ 6 callersFunctionsplit_mantle_region_prefix
(model: str)
litellm/llms/bedrock_mantle/common_utils.py:36
↓ 6 callersFunctionsplit_reasoning_tag
Split gpt-oss's inline ``<reasoning>...</reasoning>`` prefix out of a complete message. Runs the streaming splitter over the whole message,
litellm/llms/bedrock/chat/chat_completions/transformation.py:242
↓ 6 callersFunctionsplit_server_prefix_from_name
Return the unprefixed name plus the server name used as prefix. Cuts at the FIRST separator, so the two halves are only trustworthy as a pair
litellm/proxy/mcp_server/utils.py:443
↓ 6 callersFunctionsplit_sse_chunk_into_events
Split a chunk that may contain multiple SSE events into individual events. SSE format: "event: type\ndata: {...}\n\n" Multiple events in
litellm/llms/anthropic/messages/stream_assembly.py:30
↓ 6 callersFunctionsse
(payload: dict[str, object], event: str | None = None)
tests/e2e/perf/setup/mocks/reply_common.py:164
↓ 6 callersFunctionsse_response
(events: Iterable[tuple[str, Mapping[str, object]]])
tests/rust-python-harness/strategies/trace_parity/fixtures.py:31
↓ 6 callersFunctionstage_core_distribution
(source: Path, destination: Path, *, package: str = "litellm-core")
scripts/build_core_distribution.py:31
↓ 6 callersFunctionstage_private_json
Write JSON to a private temp file beside `path`, ready for `commit_staged_json`. Staging is the half that can fail on a read-only or full directo
litellm/litellm_core_utils/private_json.py:19
↓ 6 callersMethodstart
(self)
litellm/integrations/clickhouse/clickhouse_batch_logger.py:47
↓ 6 callersMethodstart
(self)
tests/integration/observability/test_signoz_delivery.py:393
↓ 6 callersMethodstart_db_health_watchdog_task
Start background tasks that monitor DB health: - A periodic SELECT 1 probe that triggers reconnect on network/connection failure. - A
litellm/proxy/utils.py:7889
↓ 6 callersMethodstart_token_refresh_task
Start the background token refresh task. This task proactively refreshes the database token before it expires, preventing co
litellm/proxy/db/prisma_client.py:742
↓ 6 callersMethodstatus
(&self)
litellm-rust/crates/gateway-inference/src/error.rs:72
↓ 6 callersMethodstop
(self)
litellm/proxy/telemetry/runtime.py:245
↓ 6 callersFunctionstop_child
(process: subprocess.Popen[bytes] | subprocess.Popen[str], grace_seconds: float = 10)
tests/integration/_support/process.py:380
↓ 6 callersFunctionstore
(gateway: Gateway, key: str, identity: str, values: Mapping[str, str])
tests/integration/mcp/test_mcp_user_env_vars.py:65
↓ 6 callersMethodstore
(&self)
litellm-rust/crates/cache-disk/tests/cache.rs:31
↓ 6 callersMethodstore_unified_file_id
(self, **kwargs)
tests/unit/proxy/openai_files_endpoint/test_storage_backend_service.py:54
↓ 6 callersFunctionstore_user_credential
Store a user credential for a BYOK MCP server.
litellm/proxy/mcp_server/db.py:1741
↓ 6 callersFunctionstored_files
(directory: Path)
tests/unit/llms/base_llm/auth/test_shared_token_store.py:82
↓ 6 callersFunctionstored_user_ids
(identity: str)
tests/integration/mcp/test_mcp_user_env_vars.py:130
↓ 6 callersFunctionstr_or_none
(value: object)
litellm/proxy/guardrails/guardrail_hooks/ovalix/ovalix_extraction.py:27
↓ 6 callersFunctionstream
()
tests/unit/proxy/hooks/test_parallel_request_limiter_v3.py:5563
↓ 6 callersFunctionstreaming_call
(chunks: Vec<Result<usize, TestError>>)
litellm-rust/crates/host-native/tests/driver.rs:194
↓ 6 callersFunctionstreams
(request: Request)
tests/integration/providers/_gemini_long_decode_support.py:140
↓ 6 callersFunctionstrip_encrypted_reasoning_blocks_from_anthropic_messages
Drop thinking / redacted_thinking blocks that carry another provider's encrypted reasoning (a turn the Responses API bridge served) before th
litellm/llms/anthropic/common_utils.py:2067
↓ 6 callersMethodstrip_encrypted_reasoning_from_input
Drop reasoning items the routed deployment cannot decrypt, keeping their readable summary. Mutates ``request_input`` in place: the router's f
litellm/responses/utils.py:669
↓ 6 callersMethodsubmit
Queue ``processor`` for closing, or hand it off once the pool is retired. The check and the put share one lock. Reading a closed flag on its
litellm/integrations/otel/plumbing/providers.py:313
↓ 6 callersMethodsubmit
( self, upload: PluginPublication, submitted_by: str, base: PluginPublicationBase | None )
litellm/repositories/skill_submission_repository.py:209
↓ 6 callersMethodsubmit
(self, fn, /, *args, **kwargs)
tests/unit/llms/anthropic/test_anthropic_wif.py:99
↓ 6 callersFunctionsuggest
( language: QueryLanguage<F>, text: string, cursor: number, lookup: (field: F) => FieldValues, )
ui/litellm-dashboard/src/components/shared/search/suggestions.ts:121
↓ 6 callersFunctionsummarize
(samples: tuple[Sample, ...])
tests/e2e/perf/setup/ui_journeys.py:70
↓ 6 callersFunctionsummarize_workspace
(data: ObservedData, connections: tuple[StoredConnection, ...])
litellm/proxy/roi_calculator/observed_workspace.py:44
↓ 6 callersMethodsunk
(self)
tests/integration/mcp/test_mcp_accounting_guardrails.py:311
↓ 6 callersFunctionsupported_prediction_headers
(headers: Mapping[str, str])
litellm/llms/anthropic/prompt_cache_prediction.py:716
↓ 6 callersMethodsupports_audio_input
(self, model: str, custom_llm_provider: str, deployment_flag: bool | None)
litellm/llms/openai/decisions/transformation.py:361
↓ 6 callersFunctionsupports_bedrock_prompt_cache_ttl
Whether Bedrock accepts a ``ttl`` ("5m"/"1h") on prompt ``cache_control`` for this model, decided by ``cache_creation_input_token_cost_above_
litellm/llms/bedrock/common_utils.py:1772
↓ 6 callersFunctionswitchToWorkerUrl
(workerUrl: string | null)
ui/litellm-dashboard/src/components/networking.tsx:270
↓ 6 callersFunctionsync_access_groups_for_deleted_model
( prisma_client: object, *, model_id: str, model_name: str, llm_router: Router | None, )
litellm/proxy/management_helpers/access_group_model_sync.py:118
↓ 6 callersMethodsync_deployment_callback_on_success
Count a success this minute; returns the counter's key.
litellm/router_rust/surface/health.py:491
↓ 6 callersFunctionsystem_clock
()
litellm-rust/crates/redis-store/src/memory.rs:13
↓ 6 callersFunctionsystemone_choice_key
(value: str | bool)
litellm/types/decisions.py:327
↓ 6 callersFunctiont
litellm/proxy/_experimental/out/_next/static/chunks/2j7hjtph8n74b.js:36
↓ 6 callersFunctiontO
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 6 callersFunctiontag_registry_cache_key
Cache key for the set of tag names that exist in ``LiteLLM_TagTable``.
litellm/proxy/common_utils/user_api_key_cache.py:526
↓ 6 callersMethodtag_request
Attach key and model tags to the active Datadog APM span. Tags set (all best-effort, skipped when value is absent): - ``lite
litellm/proxy/dd_span_tagger.py:28
↓ 6 callersMethodtake_outcome
The wrapper reports this attempt's outcome when its stream ends.
litellm/router_rust/invoke/streaming.py:201
↓ 6 callersFunctiontarget
(provider: Provider, stream: bool, continued: bool = False)
tests/integration/providers/_gemini_long_decode_support.py:200
↓ 6 callersFunctionteamMemberBudgetQueryKey
(teamId: string)
ui/litellm-dashboard/src/app/(dashboard)/hooks/teams/useTeamMemberBudgets.ts:12
↓ 6 callersFunctionteam_admin_may_edit_member_key_budgets
(general_settings: Mapping[str, object])
litellm/proxy/management_endpoints/team_admin_field_permissions.py:197
↓ 6 callersFunctionteam_admin_may_manage_projects
(general_settings: Mapping[str, object])
litellm/proxy/management_endpoints/team_admin_field_permissions.py:191
↓ 6 callersFunctionteam_model_add
Add models to a team's allowed model list. Only proxy admin or team admin can add models. Parameters: - team_id: str - Required. The tea
litellm/proxy/management_endpoints/team_endpoints.py:6845
↓ 6 callersMethodteam_named
`_team_deployments_named(team_id, name)`: the deployments the team index holds for the pair, or `None` when it does not index it.
litellm-rust/crates/router-core/src/registry.rs:766
↓ 6 callersMethodtenant_logging
(self, endpoint: str | None, key: str | None)
tests/integration/observability/test_signoz_delivery.py:350
↓ 6 callersMethodtenant_tag
(self, value: str)
tests/integration/spend/test_baggage_tag_mappings.py:69
↓ 6 callersFunctiontm
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 6 callersMethodtoAncestorLineage
(s)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 6 callersFunctionto_naive_utc
LiteLLM_BudgetWindowSpend.window_start is TIMESTAMP(3), which holds naive UTC.
litellm/proxy/db/db_transaction_queue/window_spend_update_queue.py:73
↓ 6 callersFunctionto_subject
Map v1's authenticated principal onto the resolver's Subject. tenant_id / subject_id are empty for an unauthenticated caller; the per-user arms m
litellm/proxy/mcp_server/outbound_credentials/adapter.py:65
↓ 6 callersMethodtobytes
(self)
litellm/llms/nvidia_riva/audio_transcription/audio_utils.py:46
↓ 6 callersMethodtool_names_for_server
(self, server_id: str)
tests/e2e/mcp/mcp_client.py:81
↓ 6 callersFunctiontool_result_text_indices
Positions in ``texts`` that hold content already submitted under the TOOL policy. The chat-completions guardrail flow builds ``texts`` and ``stru
litellm/proxy/guardrails/guardrail_hooks/ovalix/ovalix_extraction.py:389
↓ 6 callersFunctiontool_use
The role-TOOL contentStart/toolUse/contentEnd triple Bedrock emits per tool call.
tests/integration/_support/bedrock_realtime_peer.py:94
↓ 6 callersFunctiontrace_attributes
Trace-level fields ride on an observation's span as ``langfuse.trace.*`` style attributes in v4. On the root observation they define the trace; o
litellm/integrations/langfuse/langfuse_sdk.py:211
↓ 6 callersFunctiontranscribe_owned_start_request
( request_body: Mapping[str, object], user_api_key_dict: UserAPIKeyAuth )
litellm/proxy/pass_through_endpoints/llm_provider_handlers/transcribe_passthrough_logging_handler.py:307
↓ 6 callersMethodtransform_container_file_list_request
Transform the container file list request for OpenAI API. OpenAI API expects the following request: - GET /v1/containers/{container_i
litellm/llms/openai/containers/transformation.py:282
↓ 6 callersMethodtransform_create_file_request
Transform file request and return a pre-signed request for S3. This keeps the HTTP handler clean by doing all the signing here.
litellm/llms/bedrock/files/transformation.py:1257
↓ 6 callersMethodtransform_create_file_request
2 Cases: 1. Handle basic file upload 2. Handle batch file upload (.jsonl), staged to a temp file and uploaded in a
litellm/llms/vertex_ai/files/transformation.py:1327
↓ 6 callersMethodtransform_create_file_response
( self, model: str | None, raw_response: httpx.Response, logging_obj: LiteLLML
litellm/llms/xai/files/transformation.py:161
↓ 6 callersMethodtransform_create_vector_store_response
(self, response: httpx.Response)
litellm/llms/base_llm/vector_store/transformation.py:206
↓ 6 callersMethodtransform_generate_content_response
Transform the raw response from the generate content API. Args: model: The model name raw_response: Raw HTTP
litellm/llms/gemini/google_genai/transformation.py:374
↓ 6 callersMethodtransform_litellm_team_to_scim_group
( team: LiteLLM_TeamTable | dict, )
litellm/proxy/management_endpoints/scim/scim_transformations.py:159
↓ 6 callersFunctiontransform_mcp_tool_to_openai_tool
Convert an MCP tool to an OpenAI tool.
litellm/mcp_client/tools.py:30
↓ 6 callersMethodtransform_realtime_response
( self, message: str | bytes, model: str, logging_obj: LiteLLMLoggingObj,
litellm/llms/soniox/realtime/transformation.py:545
↓ 6 callersMethodtransform_request
Transform the overall request to be sent to the API. For magistral models, adds reasoning system prompt when reasoning_effort is spec
litellm/llms/mistral/chat/transformation.py:948
↓ 6 callersMethodtransform_response
( self, response_list: list[dict[str, object]], # mutable-ok: mirrors override contract
litellm/llms/bedrock/embed/amazon_nova_transformation.py:329
↓ 6 callersMethodtransform_response
Transform the Clarifai response to a standard ModelResponse. Since Clarifai is OpenAI-compatible, we use OpenAI response transformati
litellm/llms/clarifai/chat/transformation.py:91
↓ 6 callersMethodtransform_response_api_response
( self, model: str, raw_response: httpx.Response, logging_obj: "LiteLLMLogging
litellm/llms/chatgpt/responses/transformation.py:156
↓ 6 callersMethodtransform_response_api_response
(self, **kwargs)
tests/unit/responses/test_streaming_iterator.py:2320
↓ 6 callersMethodtransform_response_dict_to_openai_response
( cls, model_response: ImageResponse, response_dict: dict )
litellm/llms/bedrock/image_generation/amazon_stability1_transformation.py:122
↓ 6 callersMethodtransform_search_vector_store_request
( self, vector_store_id: str, query: str | list[str], vector_store_search_opti
litellm/llms/bedrock/vector_stores/transformation.py:228
↓ 6 callersMethodtransform_vertex_ai_batch_list_response_to_openai_list_response
Transforms Vertex AI batch list response into OpenAI-compatible list response.
litellm/llms/vertex_ai/batches/transformation.py:257
↓ 6 callersMethodtranslate_context_management_to_responses_api
Convert Anthropic context_management dict to OpenAI Responses API array format. Anthropic format: {"edits": [{"type": "compact_20260
litellm/llms/anthropic/messages/responses_adapters/transformation.py:508
↓ 6 callersMethodtranslate_openai_usage_to_anthropic_usage_delta
(cls, usage: Usage)
litellm/llms/anthropic/messages/adapters/transformation.py:1711
↓ 6 callersFunctiontruncate_fallback_error_detail
Bound a fallback failure detail before it is logged or appended to an exception message. Each level of the fallback walk records the failure
litellm/litellm_core_utils/routing_common_utils.py:171
↓ 6 callersFunctionty
(e)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 6 callersMethodtype_name
Python's type name, as it appears in `TypeError` messages.
litellm-rust/crates/python-compat/src/value.rs:26
↓ 6 callersFunctionu
()
litellm/proxy/_experimental/out/_next/static/chunks/11et800l-khs-.js:1
↓ 6 callersFunctionu
litellm/proxy/_experimental/out/_next/static/chunks/1f2vcab0yti03.js:29
↓ 6 callersFunctionuc
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:10
↓ 6 callersFunctionui_junit_properties
(samples: tuple[Sample, ...])
tests/e2e/perf/setup/ui_journeys.py:93
↓ 6 callersMethodunbilled_usage_on_session_close
(self, model: str)
litellm/llms/meta/realtime/transformation.py:512
↓ 6 callersMethodunmanaged_db_keys
(self)
litellm/proxy/config_resolvers/settings_store.py:94
↓ 6 callersFunctionunreadable
(sources: Mapping[str, str])
tests/code_coverage_tests/check_workflow_job_name_collisions.py:407
↓ 6 callersFunctionunsupported_video_fields
( params: VideoCreateOptionalRequestParams, model: str, provider: str, drop_params: bool,
litellm/llms/base_llm/videos/async_task_utils.py:24
↓ 6 callersMethoduntil
(self, predicate: Callable[[Sequence[_State]], bool])
tests/unit/test_router/test_router_callback_hook_sequence.py:168
↓ 6 callersFunctionunwrapProxyErrorMessage
(raw: string)
ui/litellm-dashboard/src/lib/http/client.ts:88
↓ 6 callersMethodupdate
( self, *, where: Mapping[str, object], data: Mapping[str, object], include: Mapping[str, bool] )
litellm/proxy/management_endpoints/model_management_endpoints.py:339
↓ 6 callersMethodupdateResult
()
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
← previousnext →13,601–13,700 of 202,479, ranked by callers