MCPcopy Create free account

hub / github.com/BerriAI/litellm / functions

Functions202,479 in github.com/BerriAI/litellm

↓ 20 callersFunctionej
litellm/proxy/_experimental/out/_next/static/chunks/1a2lzi3ursud4.js:1
↓ 20 callersFunctionencrypt_callback_vars
Return a deep copy of metadata with callback_vars values encrypted at rest. Idempotent: a value that already decrypts cleanly is left unchanged s
litellm/proxy/common_utils/callback_utils.py:722
↓ 20 callersMethodexecute
( &self, call: MessagesCall, interceptors: &impl litellm_host::interceptors::Intercept
litellm-rust/crates/inference-messages/src/lib.rs:54
↓ 20 callersMethodextract
(document: Borrowed<'_, '_, PyAny>)
litellm-rust/crates/python-bridge/src/routes/ocr/document.rs:64
↓ 20 callersFunctionfaked_model
(scenario: Scenario, wire: Wire, model: str = "openai/faked-stream-backend")
tests/integration/streaming/faked_responses_case.py:136
↓ 20 callersFunctionfinalize_response_cost
The finalized cost, or ``None`` when it cannot be known yet: the request is still served and the legacy callback path bills it, so a finalization
litellm/proxy/spend_tracking/cost_finalization.py:22
↓ 20 callersMethodfind_many
(self, **_: object)
tests/unit/proxy/test_admin_mcp.py:529
↓ 20 callersMethodformat
(self, record)
litellm-proxy-extras/litellm_proxy_extras/_logging.py:12
↓ 20 callersFunctionfunction_setup
( original_function: str, rules_obj: Rules, start_time: datetime.datetime, *args: Any, # posi
litellm/utils.py:1266
↓ 20 callersFunctionget_audit_log_changed_by
( *, litellm_changed_by: str | None, user_api_key_dict: UserAPIKeyAuth, litellm_proxy_admin_na
litellm/proxy/management_helpers/audit_logs.py:54
↓ 20 callersMethodget_base_model
(model: str)
litellm/llms/xai/common_utils.py:84
↓ 20 callersFunctionget_chain_id_from_headers
Extract chain id for call chaining from request headers. Priority order: 1. ``x-litellm-trace-id`` (explicit, highest priority) 2. `
litellm/proxy/litellm_pre_call_utils.py:1013
↓ 20 callersFunctionget_logging_caching_headers
(request_data: dict)
litellm/proxy/common_utils/callback_utils.py:481
↓ 20 callersFunctionget_model_info_helper
Helper for 'get_model_info'. Separated out to avoid infinite loop caused by returning 'supported_openai_param's
litellm/utils.py:7250
↓ 20 callersFunctionget_num_retries_from_retry_policy
Prefer NotFoundErrorRetries for any 404 answer, then walk the exception's MRO most specific class first.
litellm/litellm_core_utils/get_retry_from_policy.py:53
↓ 20 callersFunctionget_original_file_id
Extract the original provider file/batch ID from an encoded ID. Handles IDs that start with "file-" or "batch_" prefix.
litellm/proxy/openai_files_endpoints/common_utils.py:322
↓ 20 callersFunctionget_s3_object_key
( s3_path: str, prefix: str, start_time: datetime, s3_file_name: str, partition_granularit
litellm/integrations/s3.py:384
↓ 20 callersMethodget_supported_openai_params
(self, model: str)
litellm/llms/bedrock/chat/converse_transformation.py:705
↓ 20 callersFunctionget_team_object
(team_id: str, **_: object)
tests/unit/proxy/auth/test_auth_checks_object_access_and_lookup.py:5046
↓ 20 callersFunctionhandle_generic_stream_chunk
Parse a single GENERIC SSE chunk into a LiteLLM ModelResponseStream.
litellm/llms/oci/chat/generic.py:367
↓ 20 callersMethodhandle_model_invocation_job_status
Handle ``GetModelInvocationJob`` status check for AWS Bedrock bulk batch inference jobs (the ARN type returned by ``CreateModelInvoca
litellm/llms/bedrock/batches/handler.py:278
↓ 20 callersMethodhandle_streaming
Handle streaming A2A request. Args: request_id: A2A JSON-RPC request ID params: A2A MessageSendParams contai
litellm/a2a_protocol/providers/base.py:39
↓ 20 callersFunctionhealthy_json
(blob: str = ORDER_TWO_BLOB)
tests/integration/routing/test_fallback_hop_foreign_encrypted_reasoning.py:116
↓ 20 callersFunctioninfo_key_fn
Retrieve information about a key. Parameters: - key: str | None (query parameter) - The key to look up. Accepts the plaintext key or its
litellm/proxy/management_endpoints/key_management_endpoints.py:5032
↓ 20 callersFunctioninvalid
(error: serde_json::Error)
litellm-rust/crates/llms/src/base_llm/sse.rs:72
↓ 20 callersMethodis_model_gpt_5_model
(cls, model: str)
litellm/llms/openai/chat/gpt_reasoning_transformation.py:120
↓ 20 callersMethoditer_bytes
(self, chunk_size: int)
tests/unit/llms/openai/test_openai_file_content_streaming.py:275
↓ 20 callersMethodkey
(&self, suffix: &str)
litellm-rust/crates/redis-store/src/rate_limits/keys.rs:9
↓ 20 callersFunctionl
litellm/proxy/_experimental/out/_next/static/chunks/2k-fkubkt19cc.js:8
↓ 20 callersFunctionlist_passthrough_ids_from_db
Query the DB for managed IDs the caller owns and return an OpenAI-style paginated list response. Returns ``None`` when ``prisma_client`` is u
litellm/proxy/pass_through_endpoints/managed_id_rewriter.py:1131
↓ 20 callersMethodlisted
(self, key: str, params: Mapping[str, str] | None = None)
tests/integration/management/test_vector_store_file_list_managed_ids.py:410
↓ 20 callersFunctionlisted_ids
( gateway: Gateway, path: str = "/v1/models", *, key: str | None = None, params: Mapping[s
tests/integration/_support/bedrock_discovery.py:194
↓ 20 callersMethodloaded_model_cost_map
The catalog as last loaded (bundled or remote), untouched by ``register_model`` or router registrations.
litellm/litellm_core_utils/get_model_cost_map.py:103
↓ 20 callersFunctionmake_response
Build a minimal ModelResponse-like dict.
tests/unit/integrations/newrelic/test_newrelic.py:88
↓ 20 callersFunctionmaybe_settle_background_interaction_before_delete
( interaction_id: str, delete_kwargs: Mapping[str, object], fetch_interaction: FetchInteraction =
litellm/interactions/background_cost_polling.py:787
↓ 20 callersFunctionmock_generator
()
tests/unit/proxy/test_litellm_pre_call_utils.py:3527
↓ 20 callersFunctionmodelAvailableCall
( accessToken: string, userID: string, userRole: string, return_wildcard_routes: boolean = false, te
ui/litellm-dashboard/src/components/networking.tsx:1844
↓ 20 callersMethodmodel_dump
(self)
litellm/proxy/management_helpers/utils.py:56
↓ 20 callersFunctionnewKeyResourceData
(t *testing.T, raw map[string]interface{})
terraform/provider/litellm/resource_key_test.go:21
↓ 20 callersFunctionnewTeamTestServer
(t *testing.T, captured *map[string]interface{}, infoBody string)
terraform/provider/litellm/resource_team_test.go:17
↓ 20 callersMethodnow
(self)
tests/unit/proxy/db/test_log_db_metrics_service_spans.py:120
↓ 20 callersMethodnullable
()
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 20 callersMethodobserve
(self, value)
tests/unit/enterprise/integrations/test_prometheus.py:770
↓ 20 callersMethodon
(e,t,n)
litellm/proxy/_experimental/out/_next/static/chunks/3lb5gjzjxjxem.js:1
↓ 20 callersFunctionopen
( config: &HttpClientConfig, url: &str, headers: HeaderMap, )
litellm-rust/crates/http/tests/websocket.rs:116
↓ 20 callersFunctionordered_group
( scenario: Scenario, *, order_one: StoredResponse, order_two: StoredResponse, model: str
tests/integration/routing/test_fallback_hop_foreign_encrypted_reasoning.py:331
↓ 20 callersMethodorganization
(self, **fields: JsonValue)
tests/integration/_support/client.py:212
↓ 20 callersFunctionpages_response
()
litellm-rust/crates/inference-ocr/tests/ocr/main.rs:140
↓ 20 callersFunctionparse_database_endpoint
Parse a PostgreSQL DSN into its exportable endpoint identity. Returns ``None`` for an absent, malformed or non-PostgreSQL URL rather than rai
litellm/integrations/otel/model/db_endpoint.py:41
↓ 20 callersFunctionpost_chat
(gateway: Gateway, body: Mapping[str, JsonValue])
tests/integration/routing/_prompt_cache_pin_support.py:278
↓ 20 callersMethodprocess_output_streaming_response
( self, responses_so_far, guardrail_to_apply, litellm_logging_obj=None, user_api_key_dict=None, reques
tests/unit/proxy/policy_engine/test_pipeline_executor.py:1188
↓ 20 callersFunctionprompt
(marker: str)
tests/integration/_support/anthropic_thinking.py:31
↓ 20 callersFunctionquery
Query a RAG pipeline.
litellm/rag/main.py:400
↓ 20 callersMethodread
(self, file_path: str, offset: int = 0, limit: int = 2000)
litellm/llms/deepagents/harness/sandbox_backend.py:223
↓ 20 callersFunctionrecord_backend
(root: Path)
tests/e2e_harness/test_provider_edge.py:320
↓ 20 callersFunctionredact_oauth_error_body
``assertion`` may be every form of the credential that went out on the wire. A grant that encodes its credential before sending it (``client_secr
litellm/llms/base_llm/auth/token_exchange.py:123
↓ 20 callersMethodredis
(&self)
litellm-rust/crates/redis-store/src/dual_cache/mod.rs:77
↓ 20 callersFunctionregister_client_with_server
( request: Request, mcp_server: MCPServer, client_name: str, grant_types: list | None, res
litellm/proxy/mcp_server/discoverable_endpoints.py:2189
↓ 20 callersFunctionrender_soniox_tokens_as_srt
Render Soniox tokens as SRT (SubRip) subtitle format. Returns an empty string if no tokens have timestamp data.
litellm/llms/soniox/common_utils.py:159
↓ 20 callersFunctionrun_ptu_flat_cost_rollup
Rollup one UTC day of flat PTU cost across all PTU-configured model deployments. Defaults to yesterday UTC. It upserts the current charges first,
litellm/proxy/spend_tracking/ptu_flat_cost_rollup.py:461
↓ 20 callersMethodrun_with_session
Open a session, run the provided coroutine, and clean up. quiet_on_error demotes the failure line to debug for callers that own the exception
litellm/mcp_client/client.py:860
↓ 20 callersMethodscrub_gateway_admission_credentials
Remove gateway-shaped credentials and the caller's admitted credential from EVERY egress header context: top-level ``Authorization`` (oauth2 +
litellm/proxy/mcp_server/auth/user_api_key_auth_mcp.py:954
↓ 20 callersFunctionseed_users
(scenario: Scenario, count: int)
tests/integration/_support/scim.py:58
↓ 20 callersMethodseek
(self, *args)
tests/unit/llms/vertex_ai/files/test_vertex_ai_files_streaming.py:235
↓ 20 callersMethodsend
(self, message: str, /)
litellm/llms/openai/live/handler.py:96
↓ 20 callersFunctionsession_keys_from_master_key
Derive the session signing key from the proxy master key. A memory-hard scrypt KDF (RFC 7914) over a session-specific domain-label salt yields a
litellm/proxy/mcp_server/outbound_credentials/session_credentials.py:54
↓ 20 callersMethodset
(e,t)
litellm/proxy/_experimental/out/_next/static/chunks/2aatum0yo4cv8.js:1
↓ 20 callersMethodsetQueryData
(e,t,s)
litellm/proxy/_experimental/out/_next/static/chunks/0tx19jeq-wskb.js:2
↓ 20 callersFunctionsetSecureItem
(key: string, value: string)
ui/litellm-dashboard/src/utils/secureStorage.ts:15
↓ 20 callersFunctionset_provider_response_headers_in_hidden_params
( response: _CarriesHiddenParams, headers: httpx.Headers | Mapping[str, str] )
litellm/litellm_core_utils/core_helpers.py:1182
↓ 20 callersFunctionsettling_reads
(status: str, count: int)
tests/integration/spend/_background_responses.py:217
↓ 20 callersFunctionstart
(router: &Router, model: &str)
litellm-rust/crates/python-bridge/src/router/host.rs:1164
↓ 20 callersMethodsuccess_handler
(self, result=None, start_time=None, end_time=None, cache_hit=None, **kwargs)
tests/unit/litellm_core_utils/test_redact_messages.py:1767
↓ 20 callersFunctiontenants
(scenario: Scenario, wire: Wire)
tests/integration/authorization/_responses_compact.py:228
↓ 20 callersFunctiontext_completion
( prompt: str | list[str | list[str | list[int]]], # Required: The prompt(s) to generate completions for.
litellm/main.py:8056
↓ 20 callersMethodtier_names
The active tier names: the defined names, or the built-in set in severity order.
litellm/router_strategy/complexity_router/config.py:2060
↓ 20 callersFunctiontimed
(send: Callable[[], httpx.Response])
tests/integration/management/_batch_file_caps.py:248
↓ 20 callersFunctiontoken_count_log_lines
The structured fields of every INFO token count line the proxy logger writes while the test runs.
tests/unit/proxy/conftest.py:484
↓ 20 callersMethodtransform_image_edit_response
(self, model, raw_response, logging_obj)
tests/unit/llms/custom_httpx/test_llm_http_handler.py:5051
↓ 20 callersMethodtransform_realtime_request
( self, message: str, model: str, session_configuration_request: str | None =
litellm/llms/soniox/realtime/transformation.py:501
↓ 20 callersMethodtransform_search_request
Transform a LiteLLM search request to TinyFish's querystring format. Maps LiteLLM's unified-spec params (see ``BaseSearchCon
litellm/llms/tinyfish/search/transformation.py:85
↓ 20 callersFunctiontwo_deployments
( **params: object, )
tests/unit/test_router/core/test_retries_cooldowns.py:87
↓ 20 callersMethodunmount
()
litellm/proxy/_experimental/out/_next/static/chunks/0tx19jeq-wskb.js:2
↓ 20 callersMethodunshift
(e)
litellm/proxy/_experimental/out/_next/static/chunks/14v3irtjifjt2.js:1
↓ 20 callersMethodupdate_database
Record the request's spend, answering whether its cost still needs charging. False only for a batch or video poll whose cost row another poll
litellm/proxy/db/db_spend_update_writer.py:704
↓ 20 callersFunctionupdate_spend_logs_job
Job to process spend_log_transactions queue. This job is triggered based on queue size rather than time. Pops the batch once, writes spe
litellm/proxy/utils.py:9054
↓ 20 callersFunctionupstream
Starts an upstream that answers its n-th request with the n-th response and 404s after.
litellm-rust/crates/inference-chat/tests/support/mod.rs:28
↓ 20 callersMethodupstream_tools
(self)
tests/integration/mcp/test_mcp_llm_endpoints.py:549
↓ 20 callersFunctionuseComboboxAnchor
()
ui/litellm-dashboard/src/components/ui/combobox.tsx:256
↓ 20 callersFunctionuser_api_key_auth
(request: Request, api_key: str)
tests/unit/proxy/test_configs/custom_auth.py:9
↓ 20 callersMethodvalidate_environment
Expects LAGO_API_BASE, LAGO_API_KEY, LAGO_API_EVENT_CODE, Optional: LAGO_API_CHARGE_BY in t
litellm/integrations/lago.py:43
↓ 19 callersFunctionGetStringValue
Helper functions to handle potential nil values from the API response
terraform/provider/litellm/utils.go:118
↓ 19 callersFunctionQ
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
↓ 19 callersFunctionS
(e)
litellm/proxy/_experimental/out/_next/static/chunks/2_9r2jeem0dh8.js:1
↓ 19 callersMethod_aliased_server
(**overrides: Any)
tests/unit/proxy/mcp_server/test_mcp_server_manager.py:13628
↓ 19 callersFunction_allow_response
(correlation_id: str = "corr-1")
tests/unit/proxy/guardrails/guardrail_hooks/test_agent_365.py:52
↓ 19 callersFunction_always
(reply: Reply)
tests/integration/messages_endpoint/providers/anthropic/test_anthropic_messages_overloaded_wire.py:65
↓ 19 callersFunction_assert_timed_out
(status: int, text: str, seconds: float = _TIMEOUT_SECONDS)
tests/integration/providers/test_bedrock_stream_timeout_wire.py:290
↓ 19 callersFunction_assert_tool_turn
(prompt: str, result: str = _RESULT)
tests/integration/providers/test_ollama_prompt_tools_wire.py:156
↓ 19 callersFunction_await_outcome
(interaction_id: str, outcome: str, seconds: float = 30)
tests/integration/spend/test_background_interaction_settlement.py:335
← previousnext →2,801–2,900 of 202,479, ranked by callers