MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 7 callersFunctionc
(e,t,n,s)
litellm/proxy/_experimental/out/_next/static/chunks/0ayyip3c-drm1.js:1
↓ 7 callersFunctionc
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/14v3irtjifjt2.js:1
↓ 7 callersFunctioncache_byok_credential
(user_id: str, server_id: str, credential: str | None)
litellm/proxy/mcp_server/byok_credential_cache.py:34
↓ 7 callersFunctioncalculate_request_duration
Calculate audio duration from file content. Args: file: The audio file (can be file path, bytes, or file-like object) Returns:
litellm/litellm_core_utils/audio_utils/utils.py:272
↓ 7 callersFunctioncalculate_total_usage
Assume most recent usage chunk has total usage uptil then.
litellm/litellm_core_utils/streaming_handler.py:3042
↓ 7 callersFunctioncall_generation
( harness: Harness, *, body: Dict[str, Any], input_reference=None )
tests/unit/proxy/video_endpoints/test_endpoints.py:237
↓ 7 callersFunctioncall_id_of
(kwargs: Mapping[str, object])
litellm/router_strategy/least_busy.py:153
↓ 7 callersFunctioncall_metadata
(deployment_id: str = "a")
tests/unit/test_router/core/test_legacy_members.py:138
↓ 7 callersFunctioncall_next
(request)
tests/unit/proxy/mcp_server/test_mcp_oauth_passthrough_tools.py:816
↓ 7 callersMethodcall_tool
Call an MCP Tool. persistent_session runs the call inside an already-open upstream session (one upstream mcp-session-id shar
litellm/mcp_client/client.py:1156
↓ 7 callersMethodcall_type
(&self)
litellm-rust/crates/callbacks-legacy-python/src/adapter.rs:149
↓ 7 callersFunctioncall_with_native_output_fallback
( attempt: NativeOutputAttempt, fallback: NativeOutputFallbackFactory | None, model: str, stri
litellm/llms/bedrock/chat/native_output_fallback.py:211
↓ 7 callersMethodcaller
(&self)
litellm-rust/crates/gateway-auth/src/authorization.rs:89
↓ 7 callersMethodcallers
(self, rig: Rig)
tests/integration/security/_sinks.py:284
↓ 7 callersMethodcanParse
(s)
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 7 callersFunctioncanonical_model
``provider/model``, or ``None`` when the pair names no known provider. A deployment may name its vendor in the model prefix or in a separate
litellm/router_strategy/savings_baseline.py:49
↓ 7 callersFunctioncapturing
A keep-alive upstream that decodes framing like h11 and reports every request it reads.
litellm-rust/crates/gateway-forward/tests/framing.rs:16
↓ 7 callersFunctioncarried_budget_metadata
(valid_token: UserAPIKeyAuth)
litellm/proxy/spend_tracking/carried_budget_state.py:46
↓ 7 callersFunctioncarry_team_and_user_budget_state
( valid_token: UserAPIKeyAuth, team_object: LiteLLM_TeamTable | None, user_object: LiteLLM_UserTab
litellm/proxy/spend_tracking/carried_budget_state.py:18
↓ 7 callersFunctioncassette_cache_health
()
tests/_vcr_redis_persister.py:51
↓ 7 callersFunctionchat
()
tests/unit/proxy/test_route_priority.py:71
↓ 7 callersMethodchat
(self, key: str, model: str, content: str, *, max_tokens: int = 16)
tests/e2e/quota_management/ratelimit/quota_client.py:21
↓ 7 callersFunctionchat_body
(model: str, parts: Sequence[JsonValue], text: str, **fields: JsonValue)
tests/integration/_support/pdf_document.py:134
↓ 7 callersFunctionchat_completion_pass_through_endpoint
( fastapi_response: Response, request: Request, adapter_id: str, user_api_key_dict: UserAPIKey
litellm/proxy/pass_through_endpoints/pass_through_endpoints.py:273
↓ 7 callersFunctionchat_content
(frames: Sequence[Mapping[str, JsonValue]])
tests/integration/_support/responses_stream.py:134
↓ 7 callersFunctionchat_row
(custom_id: str, content: str, **body_extra: object)
tests/unit/proxy/batches_endpoints/test_litellm_executed_batches.py:62
↓ 7 callersFunctionchat_stream
(identity: str, model: str, text: str)
tests/integration/_support/openai_wire.py:43
↓ 7 callersMethodcheck
(&self)
litellm-rust/crates/cache/tests/dual.rs:423
↓ 7 callersFunctioncheck_blocked_extension
( filename: str | None, blocked_extensions: tuple[str, ...] | None, )
litellm/proxy/openai_files_endpoints/general_upload_validation.py:112
↓ 7 callersFunctioncheck_coordination_redis_connection
Test a coordination Redis connection with the provided credentials. Parameters: - settings: dict - Redis connection params to test. Cred
litellm/proxy/management_endpoints/coordination_redis_endpoints.py:407
↓ 7 callersFunctioncheck_permissions_caller_permission
Require PROXY_ADMIN when `permissions` is present in the request body. Presence is detected via `data.model_fields_set` so a caller that
litellm/proxy/management_endpoints/key_management_endpoints.py:1176
↓ 7 callersFunctionchildren
()
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:10
↓ 7 callersFunctionchunk
(delta: dict, finish: str | None)
tests/unit/llms/openai/test_openai.py:1034
↓ 7 callersMethodchunk_parser
(self, chunk: dict)
litellm/llms/bedrock/chat/chat_completions/transformation.py:284
↓ 7 callersMethodchunk_parser
(self, chunk: dict)
litellm/llms/databricks/chat/transformation.py:1074
↓ 7 callersFunctionchunked_provider
( *, chunks: tuple[bytes, ...] = SSE_CHUNKS, content_type: str = "text/event-stream", abort_af
tests/e2e_harness/test_provider_edge.py:211
↓ 7 callersFunctionclaim_affinity_pin
Return the authoritative first writer, replacing it only when it becomes ineligible. Eligible claims refresh the returned winner. Legacy deployme
litellm/caching/affinity_cache.py:146
↓ 7 callersFunctionclaimed_capabilities
The licensed capabilities this complexity config claims.
litellm/router_strategy/auto_router/auto_router_model_naming.py:336
↓ 7 callersFunctionclassifier_input_snapshot
(value: object, *, openai_sdk: bool = False)
litellm/litellm_core_utils/classifier_logging.py:16
↓ 7 callersFunctionclassifier_prompt
(classifier_request: Request)
tests/integration/_support/safeguards_rig.py:405
↓ 7 callersFunctionclassify_null_flow_row
( *, has_per_user_tokens: bool, authorization_url: str | None, registration_url: str | None,
litellm/proxy/mcp_server/oauth2_flow_backfill.py:133
↓ 7 callersFunctionclaude_deployment
(scenario: Scenario, provider: Provider, wire_url: str, upstream_url: str)
tests/integration/providers/test_chat_completions_tool_search_wire.py:476
↓ 7 callersFunctionclean_competitor_line
Strip numbering, bullets, and whitespace from a competitor name line.
litellm/proxy/management_endpoints/policy_endpoints/endpoints.py:923
↓ 7 callersFunctionclear_pre_routing_selection
Drop any selection the router did not make itself on this hop. The buckets carry whatever the caller sent, so an inbound value is the caller
litellm/router_utils/fallback_event_handlers.py:334
↓ 7 callersMethodclose
(self)
litellm/rust_bridge/execution.py:41
↓ 7 callersFunctioncodex_model_sync_args
`-c model_catalog_json=...` pointing Codex at the proxy's model list, or why it was skipped. Codex has no env or inline equivalent of OPENCODE_CO
litellm/proxy/client/cli/commands/agents.py:873
↓ 7 callersFunctioncoding_agent_stream_message
( tool_ids: tuple[str, str], *, arguments: tuple[Mapping[str, JsonValue], Mapping[str, JsonValue]]
tests/integration/translation/chat_completions/templates.py:10
↓ 7 callersFunctioncollect
(handler, ctx, prompt)
tests/unit/llms/codex/harness/test_transformation.py:194
↓ 7 callersMethodcompletions
Create a chat completion. Args: model (str): The model to use for completion messages (List[Dict[str, str]])
litellm/proxy/client/chat.py:52
↓ 7 callersFunctionconfigure_claude_settings
Persistently route Claude Code through base_url, recording how to undo it. Both files are staged before either is committed, so a full disk or a
litellm/proxy/client/cli/commands/claude_settings.py:521
↓ 7 callersFunctionconfigured
(snapshot: &Snapshot<'_>)
litellm-rust/crates/python-bridge/src/http.rs:190
↓ 7 callersFunctionconfigured_pass_through_routes_first
( routes: Sequence[BaseRoute], builtin_routes: Collection[BaseRoute] )
litellm/proxy/route_priority.py:53
↓ 7 callersFunctionconnect
( client: &Client, url: &str, headers: HeaderMap, timeout: Option<Duration>, )
litellm-rust/crates/http/src/websocket.rs:27
↓ 7 callersMethodconnect
(self, timeout: int | timedelta | None = None)
litellm/proxy/db/routing_prisma_wrapper.py:184
↓ 7 callersFunctionconnect_guarded
( deployment: &RedisDeployment, namespace: &str, timeout: Duration, breaker: Option<Breaker>,
litellm-rust/crates/redis-client/tests/support/mod.rs:43
↓ 7 callersFunctionconnection_id
(provider: str, api_url: str)
litellm/proxy/roi_calculator/settings.py:20
↓ 7 callersFunctionconnection_info
( target: &Target, credentials: &Credentials, keepalive: bool, )
litellm-rust/crates/redis-client/src/config.rs:259
↓ 7 callersFunctioncontent_of_first_choice
(response: httpx.Response)
tests/integration/caching/response_cache_case.py:382
↓ 7 callersFunctioncontent_parts
(message: Mapping[str, object])
litellm/litellm_core_utils/prompt_templates/common_utils.py:474
↓ 7 callersFunctioncontext
(server: Option<&str>)
litellm-rust/crates/gateway-mcp/tests/gateway/support.rs:188
↓ 7 callersFunctioncontext_management_betas
(context_management: JsonValue)
litellm/llms/anthropic/common_utils.py:296
↓ 7 callersFunctioncontrol_plane_host
(region: str)
tests/integration/_support/bedrock_discovery.py:48
↓ 7 callersMethodconverse_stream
(self, guardrail: str, marker: str)
tests/integration/observability/test_bedrock_converse_stream_post_call_guardrail.py:219
↓ 7 callersFunctionconvert_key_logging_metadata_to_callback
( data: AddTeamCallback, team_callback_settings_obj: TeamCallbackMetadata | None, )
litellm/proxy/litellm_pre_call_utils.py:1185
↓ 7 callersMethodconvert_to_bedrock_format
Convert the litellm messages/response to the bedrock request format. If source is "INPUT", then messages is required. If sou
litellm/proxy/guardrails/guardrail_hooks/bedrock_guardrails.py:498
↓ 7 callersFunctioncooled
(gateway: Gateway, upstream: Upstream, group: str)
tests/integration/routing/test_priority_scheduler_queue_cleanup.py:399
↓ 7 callersFunctioncooled
(cache: OwnedRedis, deployment_id: str)
tests/integration/routing/test_payment_required_cooldown.py:83
↓ 7 callersFunctioncosine_similarity
(left: Vector, right: Vector)
litellm/proxy/common_utils/semantic_text_index.py:75
↓ 7 callersFunctioncost_per_google_maps_grounding_request
Calculates the cost of Grounding with Google Maps. Billing follows ``web_search_billing_unit`` in model_info the same way Google Search grou
litellm/llms/gemini/cost_calculator.py:86
↓ 7 callersMethodcould_close_json
Whether the buffer's logical last non-whitespace byte is "}" or "]", i.e. whether a JSON value could plausibly be complete. Tracked
litellm/litellm_core_utils/json_fragment_accumulator.py:37
↓ 7 callersMethodcount
(body)
tests/unit/proxy/anthropic_endpoints/test_endpoints.py:967
↓ 7 callersMethodcount
(self, *args, **kwargs)
tests/unit/proxy/spend_tracking/test_spend_management_endpoints.py:228
↓ 7 callersMethodcount_external_cache_breakpoints
Client breakpoints outside messages and system that the provider cap still counts. A tool carries its mark at the top level (Anthropic shape)
litellm/integrations/anthropic_cache_control_hook.py:1001
↓ 7 callersFunctioncount_prompt_tokens
( model: str, api_key: str, body: Mapping[str, JsonValue], api_base: str | None = None, )
litellm/llms/anthropic/prompt_cache_prediction.py:541
↓ 7 callersMethodcount_request
Mirrors the host's key precedence: `messages`, then `prompt`, then `input`, then `query` plus `documents`.
litellm-rust/crates/token-counter/src/counter.rs:45
↓ 7 callersFunctioncount_request_input_tokens
(model: str, request_input: object, system_input: object)
litellm/proxy/utils.py:1208
↓ 7 callersMethodcount_text
(&self, text: &str)
litellm-rust/crates/token-counter/src/counter.rs:39
↓ 7 callersFunctioncount_tokens
``_count_pre_call_check_tokens``: messages, or Responses input normalized to messages, plus the tools and Anthropic ``system`` block the message l
litellm/router_rust/steps/precall.py:69
↓ 7 callersMethodcount_tokens
( self, model_to_use: str, messages: list[dict[str, object]] | None, contents:
litellm/llms/vertex_ai/common_utils.py:1444
↓ 7 callersFunctioncount_tokens_headers_for
(api_key: str)
tests/unit/llms/anthropic/test_count_tokens_oauth.py:48
↓ 7 callersFunctioncounter_key
(counter: RateLimitCounter)
tests/unit/redis/stores/fake_rate_limits.py:44
↓ 7 callersMethodcreate
(self, *, data: Mapping[str, object])
litellm/proxy/management_helpers/utils.py:78
↓ 7 callersMethodcreate
( self, role: ActorRole, *, tenants: tuple[Tenant, ...] = (), profile: ActorProfile = "database_role"
tests/e2e/management/jwt_actors.py:126
↓ 7 callersMethodcreate_async_transport
- Creates a transport for httpx.AsyncClient - if litellm.force_ipv4 is True, it will return AsyncHTTPTransport with local_address
litellm/llms/custom_httpx/http_handler.py:1321
↓ 7 callersFunctioncreate_container
( name: str, expires_after: Mapping[str, object] | None = None, file_ids: list[str] | None = None,
litellm/containers/main.py:130
↓ 7 callersFunctioncreate_mock_client_factory
Factory function that creates mock client functions based on configuration. Returns: tuple: (create_mock_client_func, should_use_moc
litellm/integrations/mock_client_factory.py:108
↓ 7 callersFunctioncreate_mock_proxy_logging
()
tests/unit/proxy/test_update_spend.py:64
↓ 7 callersMethodcreate_team_member_budget_table
Create team member budget table with provided limits. The team's own reset period is only inherited when the caller left the member d
litellm/proxy/management_endpoints/team_endpoints.py:711
↓ 7 callersMethodcreate_user
(self, body: UserNewBody)
tests/e2e/management/management_client.py:483
↓ 7 callersMethodcreate_vertex_url
Return the base url for the vertex partner models
litellm/llms/vertex_ai/vertex_llm_base.py:320
↓ 7 callersFunctioncredential
(index: usize, access_token: bool)
litellm-rust/crates/auth-types/tests/token.rs:16
↓ 7 callersFunctioncredential_binding_matches
( binding: MCPOAuthIdentityBinding, user_id: str, server_id: str, credential: Mapping[str, obj
litellm/proxy/mcp_server/oauth_identity_binding.py:243
↓ 7 callersFunctiond
(t,e)
litellm/proxy/_experimental/out/_next/static/chunks/0cz1d0mv5g_q7.js:1
↓ 7 callersFunctiondailyActivityQuery
( entity: DailyActivityEntity, req: DailyActivityRequest, extra: QueryParams = {}, )
ui/litellm-dashboard/src/components/networking.tsx:1367
↓ 7 callersMethoddatabase_unavailable_message
The 503 detail for a service-unavailable database failure: retry guidance for a transient outage, a pointer at the deployment for a fa
litellm/proxy/db/exception_handler.py:519
↓ 7 callersFunctiondecode
Decodes a whole stream at once. An event the stream ends inside is dropped.
litellm-rust/crates/llms/src/base_llm/sse.rs:53
↓ 7 callersMethoddecode
(e)
litellm/proxy/_experimental/out/_next/static/chunks/1f2vcab0yti03.js:2
↓ 7 callersFunctiondecode_and_normalize_response
( model: &str, raw_response: &[u8], request_format: OcrResponseFormat, normalize: impl FnOnce(
litellm-rust/crates/llms/src/base_llm/ocr/transformation.rs:425
↓ 7 callersFunctiondecode_request
(wire: OcrWireRequest)
litellm-rust/crates/inference-ocr/src/wire.rs:58
↓ 7 callersFunctiondecoded_answer_object
(text: str)
litellm/llms/base_llm/base_utils.py:240
← previousnext →10,701–10,800 of 202,479, ranked by callers