MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 6 callersFunctionis_gcs_upload
(request: Request)
tests/integration/providers/_vertex_batch_rows.py:45
↓ 6 callersFunctionis_gpt_reasoning_series_name
(model: str)
litellm/llms/openai/chat/gpt_reasoning_transformation.py:100
↓ 6 callersMethodis_interactions_usage_object
(usage_object: object)
litellm/litellm_core_utils/llm_cost_calc/usage_object_transformation.py:125
↓ 6 callersFunctionis_interception_internal_key
( key: str, prefixes: frozenset[str] = INTERCEPTION_INTERNAL_PREFIXES, )
litellm/types/integrations/custom_logger.py:39
↓ 6 callersMethodis_mcp_server_used
(self, mcp_servers: list[AnthropicMcpServerTool] | None)
litellm/llms/anthropic/common_utils.py:789
↓ 6 callersMethodis_openai_responses_route
Check if the URL route is an OpenAI responses API endpoint.
litellm/proxy/pass_through_endpoints/llm_provider_handlers/openai_passthrough_logging_handler.py:255
↓ 6 callersFunctionis_passthrough_request_streaming
Returns True if the request is streaming. A JSON body need not be an object, so a list or scalar can reach here; it carries no streaming
litellm/proxy/pass_through_endpoints/llm_passthrough_endpoints.py:244
↓ 6 callersFunctionis_past_deprecation_date
(value: Mapping[str, object])
litellm/litellm_core_utils/model_retirement.py:8
↓ 6 callersMethodis_pdf_used
Set to true if media passed into messages.
litellm/llms/anthropic/common_utils.py:814
↓ 6 callersMethodis_polling_id
Check if a response_id is a polling ID
litellm/proxy/response_polling/polling_handler.py:42
↓ 6 callersFunctionis_projected_spend_over_limit
( current_spend: float, soft_budget_limit: float | None, budget_duration: str | None = None, b
litellm/proxy/utils.py:9549
↓ 6 callersFunctionis_running
(pid: int)
litellm/proxy/client/cli/commands/autoroute/process.py:141
↓ 6 callersFunctionis_session_refresh_token
Cheap prefix check for a session REFRESH token so the token endpoint can route a refresh grant without crypto.
litellm/proxy/mcp_server/outbound_credentials/session_token.py:330
↓ 6 callersMethodis_single
The name resolved to one deployment that is served as is: a deployment id, or the default deployment (`_common_checks_available_deployment` returns a
litellm-rust/crates/router-core/src/registry.rs:345
↓ 6 callersFunctionis_str_keyed_dict
( value: object, )
litellm/litellm_core_utils/llm_cost_calc/guardrail_cost.py:130
↓ 6 callersFunctionis_string_batch_input
Return True when the only inspected content is an ``input`` list of plain strings, the /embeddings batch shape, which :func:`apply_redacted_messag
litellm/proxy/guardrails/_content_utils.py:477
↓ 6 callersMethodis_timeout
(&self)
litellm-rust/crates/http/src/error.rs:420
↓ 6 callersMethodis_token_expired
Check if the token in the given URL is expired.
litellm/proxy/db/prisma_client.py:509
↓ 6 callersMethodis_vertex_partner_model
Check if the model string is a Vertex AI Partner Model Only use this once you have confirmed that custom_llm_provider is vertex_ai
litellm/llms/vertex_ai/vertex_ai_partner_models/main.py:51
↓ 6 callersFunctionis_window_active
(window: ModelAccessWindow, now: datetime)
litellm/litellm_core_utils/access_windows.py:35
↓ 6 callersFunctionissue
(overrides: Partial<IssueForClassification> = {})
scripts/classify-issue.test.ts:53
↓ 6 callersFunctioniteratorDone
()
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 6 callersMethodjson
(self)
litellm/proxy/guardrails/guardrail_hooks/compresr/compresr.py:151
↓ 6 callersMethodjson
(self)
litellm/llms/vertex_ai/image_edit/vertex_gemini_transformation.py:36
↓ 6 callersFunctionjson_response
(body: Value)
litellm-rust/crates/inference-responses/tests/support/mod.rs:62
↓ 6 callersFunctionjson_string_leaves
Depth-first, deterministically ordered string leaves of a JSON value. Returns ``None`` when the value is nested past ``MAX_STRUCTURED_CONTENT_SCA
litellm/proxy/mcp_server/utils.py:840
↓ 6 callersFunctionjunit_properties
One numeric property per observed metric and per limit, plus a `_breakdown` JSON list of the top contributors for every metric that has slices, al
tests/e2e/perf/setup/observe.py:616
↓ 6 callersFunctionk
litellm/proxy/_experimental/out/_next/static/chunks/1r_-m9nqfar7y.js:1
↓ 6 callersFunctionk
()
litellm/proxy/_experimental/out/_next/static/chunks/1f2vcab0yti03.js:57
↓ 6 callersMethodkey
Create a virtual key; delete it on teardown. `models` restricts which models the key may call (None/[] means all). `user_id` is required for
tests/e2e/lifecycle.py:71
↓ 6 callersFunctionkeyListCall
( accessToken: string, organizationID: string | null, teamID: string | null, selectedKeyAlias: string
ui/litellm-dashboard/src/components/networking.tsx:2176
↓ 6 callersMethodkey_info_as
(self, key: str, *, caller_key: str | None = None)
tests/e2e/management/management_client.py:202
↓ 6 callersFunctionlI
(e)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 6 callersFunctionlabelText
(item: MenuItem)
ui/litellm-dashboard/src/components/leftnav.tsx:571
↓ 6 callersMethodlabels
(self, reason: str)
litellm/proxy/middleware/admission_control_middleware.py:59
↓ 6 callersFunctionlangfuse_callback_variables
(sinks: SpanSinks)
tests/integration/_support/otlp_sink.py:404
↓ 6 callersFunctionlangfuse_client_init
Initialize the Langfuse REST client with caching to prevent multiple initializations. Args: langfuse_public_key (str, optional): Pub
litellm/integrations/langfuse/langfuse_prompt_management.py:52
↓ 6 callersMethodlatency_pairs
Each first-attempt request with its value of one latency metric. `gateway_added_seconds` is the seconds the gateway added (total time minus th
tests/e2e/perf/setup/observe.py:196
↓ 6 callersFunctionli
(e)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 6 callersMethodlimit_in
(self, limits: Mapping[str, int] | None)
litellm/proxy/hooks/parallel_request_limiter_v3.py:161
↓ 6 callersMethodlist_resources
(self, *, raise_on_error: bool = False)
litellm/mcp_client/client.py:1427
↓ 6 callersMethodlist_skills
( limit: int = 20, offset: int = 0, user_api_key_dict: UserAPIKeyAuth | None = None,
litellm/llms/litellm_proxy/skills/handler.py:114
↓ 6 callersMethodlist_tools
List available tools from the server. Args: raise_on_error: When True, re-raise exceptions instead of returning a
litellm/mcp_client/client.py:1073
↓ 6 callersMethodlist_tools
List all registered tools
litellm/proxy/mcp_server/tool_registry.py:82
↓ 6 callersFunctionlisted_ids
(gateway: Gateway)
tests/integration/management/test_vector_store_config_ownership.py:40
↓ 6 callersFunctionlisten
(client: &RedisClient, channel: &str)
litellm-rust/crates/redis-client/tests/pubsub.rs:19
↓ 6 callersFunctionload
(path: &Path)
litellm-rust/crates/config/src/includes.rs:50
↓ 6 callersMethodload_credential_list
Load the credential list from the database
litellm/proxy/proxy_server.py:6656
↓ 6 callersFunctionload_credentials_from_list
Updates kwargs with the credentials if credential_name in kwarg, then swaps an admin-configured oidc/ api_key reference for the secret it poi
litellm/utils.py:929
↓ 6 callersFunctionloaded_lazy_modules
The set of lazy feature modules whose routers are actually registered on this app (tracked by _install), empty until a feature loads or eager star
litellm/proxy/_lazy_features.py:659
↓ 6 callersFunctionlog_batch_registration_result
Report the outcome of the fire-and-forget managed object write. A create that fails is not retried by a later poll, so its cost is never tracked a
litellm/proxy/pass_through_endpoints/llm_provider_handlers/batch_attribution.py:84
↓ 6 callersFunctionlogged_event_key
(kwargs: Mapping[str, object], event_id: str)
litellm/litellm_core_utils/internal_call_metadata.py:179
↓ 6 callersFunctionlogged_thinking
(response: Mapping[str, JsonValue])
tests/integration/_support/anthropic_thinking.py:233
↓ 6 callersFunctionloggingOnlyContinueFromParams
( litellmParams: | { logging_only_scope?: string | null; logging_only_continue_on_input_
ui/litellm-dashboard/src/app/(dashboard)/guardrails/_components/guardrail_info_helpers.tsx:150
↓ 6 callersMethodlogin
(self, force: bool = False, no_browser: bool = False)
litellm/llms/xai/oauth.py:154
↓ 6 callersMethodlogin
(self)
tests/unit/llms/xai/test_xai_oauth.py:789
↓ 6 callersMethodlookup
(&self, key: &str)
litellm-rust/crates/model-catalog/src/catalog.rs:401
↓ 6 callersFunctionm
()
litellm/proxy/_experimental/out/_next/static/chunks/1a2lzi3ursud4.js:1
↓ 6 callersFunctionm
()
litellm/proxy/_experimental/out/_next/static/chunks/39v0w3tgm6n4z.js:1
↓ 6 callersFunctionm
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/218jq7ygtsrfo.js:1
↓ 6 callersFunctionm
(e,t,r)
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
↓ 6 callersFunctionmachine
(secrets: Arc<dyn SecretSource>)
litellm-rust/crates/inference-messages/tests/messages/main.rs:102
↓ 6 callersFunctionmakeOpenAIChatCompletionRequest
( chatHistory: { role: string; content: string | any[] }[], updateUI: (chunk: string, model?: string) => v
ui/litellm-dashboard/src/components/llm_calls/chat_completion.tsx:27
↓ 6 callersFunctionmake_request
Make a pre-call hook request.
tests/unit/proxy/hooks/test_dynamic_rate_limiter_v3.py:505
↓ 6 callersFunctionmake_response
(payload)
tests/unit/llms/gemini/audio_transcription/test_gemini_audio_transcription_transformation.py:61
↓ 6 callersFunctionmake_sync_call
( client: HTTPHandler | None, api_base: str, headers: dict, data: str, model: str, mes
litellm/llms/anthropic/chat/handler.py:269
↓ 6 callersMethodmake_sync_openai_chat_completion_request
Helper to: - call chat.completions.create.with_raw_response when litellm.return_response_headers is True - call chat.completi
litellm/llms/openai/openai.py:603
↓ 6 callersFunctionmanage_tracing
( enabled: bool, receiver_factory: Callable[[], TraceReceiver] | None = None, settings: Mapping[st
litellm/proxy/tracing_runtime.py:38
↓ 6 callersFunctionmanager
(server: &MockServer)
litellm-rust/crates/secrets-azure/tests/key_vault.rs:14
↓ 6 callersMethodmap_openai_context_management_to_anthropic
OpenAI format: [{"type": "compaction", "compact_threshold": 200000}] Anthropic format: { "edits": [ {
litellm/llms/anthropic/chat/transformation.py:1920
↓ 6 callersMethodmap_openai_params
( self, non_default_params: dict, # mutable-ok: BaseConfig signature optional_params:
litellm/llms/bedrock/chat/chat_completions/transformation.py:419
↓ 6 callersMethodmap_openai_params
( self, video_create_optional_params: VideoCreateOptionalRequestParams, model: str,
litellm/llms/volcengine/videos/transformation.py:556
↓ 6 callersFunctionmap_openai_size_to_gemini_image_config
(size: str, model: str)
litellm/llms/gemini/common_utils.py:122
↓ 6 callersFunctionmap_tool_json_text
(value: object, visit: Callable[[str], str])
litellm/proxy/guardrails/_content_utils.py:200
↓ 6 callersMethodmap_user_to_teams
Map user to teams. - If user is not in team, add them to the team - If user is in team, do nothing
litellm/proxy/auth/handle_jwt.py:2107
↓ 6 callersMethodmark
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 6 callersMethodmark_pre_call_hook_ran
Record that this guardrail's ``async_pre_call_hook`` already ran for this request, so the deployment-level hook does not run it a sec
litellm/integrations/custom_guardrail.py:999
↓ 6 callersFunctionmarker_of
(request: Request)
tests/integration/_support/anthropic_thinking.py:43
↓ 6 callersFunctionmarker_provider
(model: str = MODEL, stream_pause: float = 0)
tests/integration/providers/_coralbricks.py:303
↓ 6 callersMethodmarksAcross
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 6 callersFunctionmarks_gateway_injection
Whether the gateway put cache breakpoints on the payload THIS row was billed for. ``AnthropicCacheControlHook.record_gateway_injection`` stamps t
litellm/proxy/spend_tracking/savings.py:399
↓ 6 callersFunctionmatch_fill_missing_generalizations
Return flagged capability rules matching ``model`` for ``provider``. Later rules override earlier ones on key conflicts. Only rules listing `
litellm/litellm_core_utils/fallback_generalizations.py:234
↓ 6 callersMethodmax_size_in_memory
(&self)
litellm-rust/crates/cache-memory/src/cache.rs:177
↓ 6 callersMethodmaybe_inject_guardrail_auto_response_disable
Fold the transcription-guardrail auto-response disable into the setup. Gemini/Vertex Live reject a second ``setup`` (1007), so the guardrail'
litellm/litellm_core_utils/realtime_streaming.py:1591
↓ 6 callersMethodmcp
(self, name: str)
tests/integration/mcp/test_mcp_llm_endpoints.py:1726
↓ 6 callersFunctionmcp_authorize
( request: Request, server_id: str, user_api_key_dict: UserAPIKeyAuth = Depends(_mcp_o
litellm/proxy/management_endpoints/mcp_management_endpoints.py:2440
↓ 6 callersFunctionmcp_proxy_tool_id
(tool: Tool)
litellm/proxy/mcp_server/tool_search.py:225
↓ 6 callersFunctionmcp_server_display_names
server_id -> alias, falling back to server_name; config-only servers also fall back to their registry name.
litellm/proxy/management_helpers/resource_display_names.py:14
↓ 6 callersFunctionmembership_user_ids
(team: str)
tests/integration/_support/scim.py:139
↓ 6 callersFunctionmerge_bedrock_invoke_headers
Merge the ``X-Amzn-*`` headers the Invoke paths derive from params. ``caller_owned`` defers to a header the caller already set. ``proxy_owne
litellm/llms/bedrock/request_metadata.py:250
↓ 6 callersFunctionmerge_beta_headers
Python's `_merge_beta_headers`, over every casing of the header at once: the union of what the caller sent and `added` replaces the header, sorted and
litellm-rust/crates/llms/src/anthropic/common_utils.rs:208
↓ 6 callersFunctionmerged_guardrail_entries
Merges guardrail lists so each guardrail is named once: the first ``{name: settings}`` entry naming it wins over any plain name in either order, e
litellm/integrations/custom_guardrail.py:224
↓ 6 callersFunctionmessage_delta
(raw: bytes)
tests/integration/messages_endpoint/safeguards/test_safeguard_results_synthesized_streams.py:201
↓ 6 callersFunctionmessages_body
(model: str, marker: str, *, stream: bool)
tests/integration/providers/_cache_control_marks_support.py:129
↓ 6 callersFunctionmessages_body
(model: str, blocks: Sequence[JsonValue], text: str, **fields: JsonValue)
tests/integration/_support/pdf_document.py:125
↓ 6 callersMethodmessages_raw
POST /v1/messages (Anthropic-native body): raw outcome judged by status/body/headers, for tests that need x-litellm-call-id. With ``st
tests/e2e/logging/logging_client.py:530
↓ 6 callersMethodmessages_stream
( self, key: str, model: str, content: str, *, max_tokens: int )
tests/e2e/quota_management/spend_tracking/spend_e2e_client.py:294
↓ 6 callersFunctionmetadata_of
(kwargs: Mapping[str, object])
tests/unit/test_router/core/fallbacks_helpers.py:45
↓ 6 callersFunctionmh
(s,o)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 6 callersFunctionmigrate_passwords_to_scrypt_async
Migrate plaintext passwords in the DB to scrypt. SHA256 passwords are left alone (they migrate on next login via the SHA256 fallback). Sk
litellm/proxy/utils.py:8548
← previousnext →13,201–13,300 of 202,479, ranked by callers