Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/BerriAI/litellm
/ functions
Functions
202,479 in github.com/BerriAI/litellm
⨍
Functions
202,479
◇
Types & classes
26,341
↳
Endpoints
5,595
↓ 10 callers
Method
_token_exchange_server
(self, server_id: str)
tests/unit/proxy/mcp_server/test_mcp_server_manager.py:2666
↓ 10 callers
Function
_tool_chunk
(call_id: str, name: Optional[str], arguments: Optional[str])
tests/unit/llms/anthropic/messages/adapters/test_streaming_iterator_first_delta.py:68
↓ 10 callers
Function
_tool_name
(length: int)
tests/integration/messages_endpoint/test_guardrail_long_tool_names_wire.py:35
↓ 10 callers
Method
_tool_output
(items)
tests/unit/llms/anthropic/messages/responses_adapters/test_responses_adapters_transformation.py:1968
↓ 10 callers
Function
_transform
(payload: object)
tests/unit/llms/cohere/embed/test_transformation.py:21
↓ 10 callers
Method
_transform
( self, response_json: dict[str, object], request_data: dict[str, object] | None = None )
tests/unit/llms/ollama/test_ollama_completion_transformation.py:443
↓ 10 callers
Method
_transform
(self, input, params=None)
tests/unit/llms/bedrock_mantle/test_bedrock_mantle_responses_transformation.py:778
↓ 10 callers
Function
_transform_item
(item)
tests/unit/responses/litellm_completion_transformation/test_reasoning_input_item_preservation.py:26
↓ 10 callers
Method
_transform_responses_api_output_to_otel
Transform Responses API output items into OTEL GenAI 1.38 format. The Responses API returns output as a list of items, each with a
litellm/integrations/opentelemetry.py:2929
↓ 10 callers
Method
_translate
(self, tool_result_content)
tests/unit/llms/anthropic/messages/responses_adapters/test_responses_adapters_transformation.py:1964
↓ 10 callers
Function
_translate_with_metadata
(model: str, metadata: dict[str, str], custom_llm_provider: str | None)
tests/unit/llms/anthropic/messages/adapters/test_anthropic_messages_adapters_transformation.py:1301
↓ 10 callers
Method
_turn
(*parts)
tests/unit/router_strategy/test_complexity_router.py:16770
↓ 10 callers
Function
_unified_stream_guardrail
(seen: Dict[str, Any], block: bool = False)
tests/unit/proxy/utils/proxy_logging/test_guardrail_pipeline.py:2592
↓ 10 callers
Function
_update_team_as
(client: ManagementClient, caller_key: str, body: TeamSettingsUpdate)
tests/e2e/management/test_team_management_e2e.py:467
↓ 10 callers
Function
_upload
(gateway: Gateway, key: str, target_model_names: str, filename: str)
tests/integration/management/test_vector_store_file_list_managed_ids.py:292
↓ 10 callers
Function
_upstream_targets
(wire: Wire)
tests/integration/authorization/test_passthrough_route_denylist.py:37
↓ 10 callers
Method
_url
(self, path: str)
litellm/proxy/roi_calculator/github.py:381
↓ 10 callers
Function
_use_aes
Flip the write-time algorithm to AES-256-GCM for the duration of a test.
tests/unit/proxy/common_utils/test_encrypt_decrypt_utils.py:30
↓ 10 callers
Function
_verdicts
(entries: tuple[dict[str, object], ...])
tests/integration/observability/test_content_filter_logging_only.py:239
↓ 10 callers
Method
_vertex_embeddings_output_row
(self, **overrides)
tests/unit/llms/vertex_ai/files/test_vertex_ai_files_transformation.py:1675
↓ 10 callers
Function
_virtual_time_subscriber
( pubsub: _VirtualTimePubSub, clock: _FakeClock, events: List[str], resynced: asyncio.Event,
tests/unit/proxy/common_utils/test_config_sync_pubsub.py:357
↓ 10 callers
Function
_window
(budget_duration: str, max_budget: float)
tests/unit/proxy/test_proxy_server_endpoints_and_startup.py:14041
↓ 10 callers
Function
_wire_audit_log_callback
(monkeypatch: pytest.MonkeyPatch)
tests/unit/proxy/management_endpoints/test_team_endpoints.py:15089
↓ 10 callers
Function
_wire_size_probes
Answer the size pass's catalog reads: the storage probe walks `storage_rows` one measurement per call (repeating the last), the step-cutoff q
tests/unit/proxy/test_spend_log_cleanup.py:1739
↓ 10 callers
Function
_with_flag
(model_info, declared=None, enabled=True)
tests/unit/litellm_core_utils/test_ptu_pricing.py:32
↓ 10 callers
Function
_worker
(gateway: Gateway)
tests/integration/mcp/test_mcp_listed_tool_metadata.py:124
↓ 10 callers
Function
_write_config_file
(home: Path, config: dict[str, str])
tests/unit/proxy/client/cli/test_global_options.py:103
↓ 10 callers
Function
_written
(sink: ObjectSink, name: str)
tests/integration/observability/test_batch_line_items_cold_storage.py:507
↓ 10 callers
Function
a
(e)
litellm/proxy/_experimental/out/_next/static/chunks/0_6jqgnorqxl2.js:1
↓ 10 callers
Method
abort
()
litellm/proxy/_experimental/out/_next/static/chunks/2hk8n3azev--d.js:11
↓ 10 callers
Method
acquire
(self, owner: str, status: ROISyncStatus, scheduled_interval: float = 0)
tests/unit/proxy/roi_calculator/test_sync.py:562
↓ 10 callers
Method
acreate_file
(self, model: str, **kwargs: object)
litellm/router.py:278
↓ 10 callers
Function
active_sso_provider
The provider the SSO callback will dispatch to. Mirrors the callback's precedence rather than reporting everything configured: an environment
litellm/proxy/management_endpoints/sso/id_jag_assertion_capture.py:35
↓ 10 callers
Method
addToSet
(e)
litellm/proxy/_experimental/out/_next/static/chunks/085r5ieqhajig.js:1
↓ 10 callers
Function
add_guardrails_from_policy_engine
Add guardrails from the policy engine based on request context. This function: 1. Extracts "policies" from request body (if present) for
litellm/proxy/litellm_pre_call_utils.py:4363
↓ 10 callers
Method
add_request
(self, request: FlowItem)
litellm/scheduler.py:77
↓ 10 callers
Function
aggregate_token
The aggregate token verb: authorization_code and refresh_token grants for the identity-only session pair, or for the proxy-API credential when the
litellm/proxy/mcp_server/gateway_dcr_flow.py:1569
↓ 10 callers
Function
alias_listing_entries
`entries` plus one row per key or team alias whose target is listed. A listed name with no `deployed` group of its own is looked up under the mode
litellm/proxy/common_utils/model_listing_utils.py:330
↓ 10 callers
Function
alias_target
The id `GET /v1/models/{model_id}` looks `model_id` up under when one of the caller's aliases rewrites it, else None. A listed `model_id` resolves
litellm/proxy/common_utils/model_listing_utils.py:363
↓ 10 callers
Function
answered_ok
(opened: list[ByteStream] | None = None)
tests/unit/test_router/core/test_streaming_characterization_anthropic.py:63
↓ 10 callers
Function
anthropic_server_tool_calls
(message: object)
litellm/llms/anthropic/common_utils.py:2322
↓ 10 callers
Function
anthropic_sse_chunks_from_response
Re-emit a rewritten response as the SSE events of the ``source_chunks`` stream it replaces. ``message_delta`` carries the source stream's ``stop_
litellm/proxy/guardrails/anthropic_sse.py:159
↓ 10 callers
Method
apply_guardrail
Apply the Generic Guardrail API to the given inputs. This is the main method that gets called by the framework. Args:
litellm/proxy/guardrails/guardrail_hooks/generic_guardrail_api/generic_guardrail_api.py:526
↓ 10 callers
Function
apply_team_provider_credentials
Resolve credentials for a provider-only request (no model pinned) via ``get_team_provider_credentials`` and merge them into ``data`` in-place
litellm/proxy/openai_files_endpoints/common_utils.py:765
↓ 10 callers
Method
arun_code
(self, *, container, code, **kwargs)
tests/unit/integrations/code_interpreter_interception/test_handler.py:59
↓ 10 callers
Function
as_array
`value` as a JSON array, or None. See `as_object`.
litellm/proxy/guardrails/guardrail_hooks/llm_shield_proxy/payload.py:35
↓ 10 callers
Function
as_mapping
(value: object)
litellm/proxy/guardrails/guardrail_hooks/akto/akto.py:136
↓ 10 callers
Method
as_str
(&self)
litellm-rust/crates/llms-types/src/providers/anthropic/beta.rs:286
↓ 10 callers
Function
as_str_mapping
(value: object)
litellm/integrations/otel/model/utils.py:64
↓ 10 callers
Function
ask_turn
(marker: str, mark: JsonValue | None = None)
tests/integration/routing/_prompt_cache_pin_support.py:71
↓ 10 callers
Function
assert_no_hits
(hits: Sequence[Hit], context: str)
tests/integration/security/_sweeps.py:205
↓ 10 callers
Function
assert_not_delayed_by
(observed: Iterable[float], *, control: Iterable[float], injected_delay: float)
tests/integration/_support/timing.py:19
↓ 10 callers
Function
assert_openai_error
(response: httpx.Response, mapped: Mapped)
tests/integration/providers/_assistants_route_errors_support.py:241
↓ 10 callers
Function
assert_refused
(outcome: Outcome)
tests/integration/routing/test_payment_required_cooldown.py:216
↓ 10 callers
Function
assert_served_by_order_two
(answer: Answer, group: OrderedGroup)
tests/integration/routing/test_fallback_hop_foreign_encrypted_reasoning.py:633
↓ 10 callers
Function
assert_too_many_records
(response: httpx.Response, limit: int, source: str)
tests/integration/management/_batch_file_caps.py:294
↓ 10 callers
Method
async_add_cache
(self, result, dynamic_cache_object=None, **kwargs)
tests/unit/caching/test_caching_handler.py:2389
↓ 10 callers
Method
async_build_agentic_loop_plan
( self, tools: dict, model: str, messages: list[dict], response: objec
litellm/integrations/code_interpreter_interception/handler.py:470
↓ 10 callers
Method
async_get_available_deployments
( self, model_group: str, healthy_deployments: list, messages: list[dict[str,
litellm/router_strategy/lowest_latency.py:542
↓ 10 callers
Method
async_get_openai_client
( self, api_key: str | None, api_base: str | None, timeout: float | httpx.Time
litellm/llms/openai/openai.py:2421
↓ 10 callers
Method
async_increment_tokens_with_ttl_preservation
Increment token counters using Lua script to preserve existing TTL. This prevents TTL reset on every token increment. Inside a reques
litellm/proxy/hooks/parallel_request_limiter_v3.py:4206
↓ 10 callers
Function
async_io_token_reconcile_success
( dual_cache: DualCache, kwargs: Mapping[str, object] | None, response_obj: object, *, par
litellm/router_utils/pre_call_checks/io_token_rate_limit_check.py:856
↓ 10 callers
Method
async_log_success_event
(self, kwargs, response_obj, start_time, end_time)
litellm/integrations/prometheus.py:1639
↓ 10 callers
Method
async_moderation_hook
( self, data: dict, user_api_key_dict: UserAPIKeyAuth, call_type: CallTypesLit
litellm/proxy/guardrails/guardrail_hooks/lakera_ai_v2.py:673
↓ 10 callers
Method
async_post_call_streaming_iterator_hook
Process streaming response chunks. Collect content from the stream and run the bedrock OUTPUT scan (post_call only validates
litellm/proxy/guardrails/guardrail_hooks/bedrock_guardrails.py:2926
↓ 10 callers
Method
async_post_call_success_hook
( self, data: dict, user_api_key_dict: UserAPIKeyAuth, response: LLMResponseTy
litellm/proxy/guardrails/guardrail_hooks/aim/aim.py:368
↓ 10 callers
Method
async_resolve_deployment
( self, deployment: _DeploymentT, request_kwargs: Mapping[str, object] | None = None,
litellm/router_utils/kubernetes_pod_discovery.py:168
↓ 10 callers
Method
async_write_secret
( &self, name: &str, value: &SecretValue, description: Option<&str>, )
litellm-rust/crates/secrets-cyberark/src/secret_manager/write.rs:7
↓ 10 callers
Function
attach_blocking
Runs Python work that may block, such as a secret manager read or a callback that does I/O, on the runtime's blocking pool so the async workers stay f
litellm-rust/crates/host-python/src/gil.rs:75
↓ 10 callers
Function
attach_tenant_fan_out
Give ``provider`` the fan-out that delivers spans to key/team destinations. Called on the one provider published as the OTel global, and idempote
litellm/integrations/otel/plumbing/providers.py:1634
↓ 10 callers
Method
auth_check
``None`` when the keys reach a project; otherwise the reason, which is also logged. Mirrors the SDK's ``Langfuse.auth_check``: a 200 with no
litellm/integrations/langfuse/langfuse_sdk.py:1118
↓ 10 callers
Method
authenticate
( &self, credential: &SecretValue, )
litellm-rust/crates/gateway-auth/src/authentication.rs:127
↓ 10 callers
Method
auto_router_capability_limit
How many auto-routers may claim each gated classifier or customization capability: unlimited (None) only when the signed license list
litellm/proxy/auth/litellm_license.py:165
↓ 10 callers
Function
bedrock_bearer_token
(api_key: str | None)
litellm/llms/bedrock/base_aws_llm.py:134
↓ 10 callers
Function
bind_target
(result: InputRequiredResult, server: MCPServer)
litellm/proxy/mcp_server/interactions.py:98
↓ 10 callers
Function
binding
(log: &Log, op: OpScript)
litellm-rust/crates/host-python/tests/support/mod.rs:175
↓ 10 callers
Function
bold
(t: str)
litellm/setup_wizard.py:152
↓ 10 callers
Function
bounded_tx
Open an interactive transaction that outlives the statement bound it carries. prisma's default 5s transaction timeout would close it mid
litellm/proxy/db/db_transaction_queue/spend_logs_partition_manager.py:131
↓ 10 callers
Function
bridge_pairs
One parity pair per base and bridged endpoint: the same stored provider response and deployment, billed through the base's own endpoint and throug
tests/integration/cost_calculation/bridge_cost_parity/case.py:112
↓ 10 callers
Function
bug_report_issue_url
(report: BugReport)
litellm/litellm_core_utils/bug_report.py:166
↓ 10 callers
Function
buildPlaygroundHeaders
( tags?: readonly string[], customHeaders?: CustomHeaders, )
ui/litellm-dashboard/src/components/llm_calls/request_headers.ts:21
↓ 10 callers
Function
build_async_openai_client
( api_key: str | None, api_base: str | None, timeout: float | httpx.Timeout | None | NotGiven = NO
litellm/llms/openai/workload_identity.py:226
↓ 10 callers
Function
bulk_update_team_member_permissions
Append permissions to existing teams. Either pass team_ids to target specific teams, or set apply_to_all_teams=True to update every team
litellm/proxy/management_endpoints/team_endpoints.py:7172
↓ 10 callers
Function
c
()
litellm/proxy/_experimental/out/_next/static/chunks/3yrgf44a9g52e.js:1
↓ 10 callers
Function
cache_team_object
( team_id: str, team_table: LiteLLM_TeamTableCachedObj, user_api_key_cache: UserApiKeyCache, p
litellm/proxy/auth/auth_checks.py:3189
↓ 10 callers
Function
call
()
tests/integration/sdk/test_router_concurrent_deployment_delete_sdk.py:189
↓ 10 callers
Function
can_user_call_model
( model: str | list[str], llm_router: Router | None, user_object: LiteLLM_UserTable | None, ke
litellm/proxy/auth/auth_checks.py:5935
↓ 10 callers
Function
carries_content
(span: ReadableSpan)
tests/unit/integrations/otel/test_otel_v2_destinations.py:5114
↓ 10 callers
Function
catalog_with_runtime_registrations
The loaded catalog plus ``register_model`` calls, without what routers register for their deployments.
litellm/utils.py:4293
↓ 10 callers
Function
changed_key_fields
Logical field names whose stored value the key-update request would change. Same JSON-value comparison as :func:`changed_team_fields`: columns co
litellm/proxy/management_endpoints/team_admin_field_permissions.py:557
↓ 10 callers
Function
chat
(gateway: Gateway, model: str, key: str, **fields: JsonValue)
tests/integration/spend/_partial_stream.py:136
↓ 10 callers
Function
chat_body
(model: str, marker: str, **extra: JsonValue)
tests/integration/_support/responses_stream.py:143
↓ 10 callers
Function
chat_identity_fields
(identity: Mapping[str, str | None])
tests/integration/observability/_langfuse_otel_metadata_support.py:367
↓ 10 callers
Function
check
(team_id: str)
tests/unit/proxy/auth/test_auth_utils.py:2338
↓ 10 callers
Method
check_managed_file_id_access
(self, data: Dict, user_api_key_dict: UserAPIKeyAuth)
enterprise/litellm_enterprise/proxy/hooks/managed_files.py:1659
↓ 10 callers
Function
classify
(rig: Rig, key: str)
tests/integration/messages_endpoint/safeguards/test_safeguard_results_caller_checks.py:51
↓ 10 callers
Function
classify_agent_subject
( binding: AgentIdentityBinding, claims: Mapping[str, object], allowed_mode: AgentExecutionMode, )
litellm/proxy/agent_endpoints/managed_identity.py:172
↓ 10 callers
Function
classify_billable_request
Map a request path to its (category, normalized route), or None if not billable.
litellm/proxy/middleware/billable_request_metrics_middleware.py:162
↓ 10 callers
Function
claude_settings_path
The settings.json Claude Code reads: under CLAUDE_CONFIG_DIR when set, else ~/.claude/settings.json.
litellm/proxy/client/cli/commands/claude_settings.py:107
↓ 10 callers
Method
cleanup_spend_logs_by_size
The size-check run scheduled on maximum_spend_logs_size_check_interval: the size rule only.
litellm/proxy/db/db_transaction_queue/spend_log_cleanup.py:1431
← previous
next →
6,901–7,000 of 202,479, ranked by callers