Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/BerriAI/litellm
/ types & classes
Types & classes
26,341 in github.com/BerriAI/litellm
⨍
Functions
202,479
◇
Types & classes
26,341
↳
Endpoints
5,595
↓ 5,741 callers
Class
UserAPIKeyAuth
Return the row in the db
litellm/types/proxy/auth/user_api_key_auth.py:129
↓ 1,765 callers
Class
Ok
litellm/proxy/mcp_server/outbound_credentials/result.py:33
↓ 1,713 callers
Class
DualCache
DualCache is a cache implementation that updates both Redis and an in-memory cache simultaneously. When data is updated or inserted, it is wr
litellm/caching/dual_cache.py:70
↓ 1,540 callers
Class
Reply
tests/integration/_support/wire.py:66
↓ 1,445 callers
Class
Router
What type checkers see: the constructor's parameters and the endpoints' return types. The endpoints are served by the core at run time (litell
litellm/router.py:244
↓ 1,055 callers
Class
Path
litellm-rust/crates/gateway-mcp/src/request.rs:33
↓ 1,033 callers
Class
ModelResponse
litellm/types/utils.py:2402
↓ 982 callers
Class
Usage
litellm/types/utils.py:2035
↓ 803 callers
Class
MCPServer
litellm/types/mcp_server/mcp_server_manager.py:96
↓ 792 callers
Class
Error
litellm/proxy/mcp_server/outbound_credentials/result.py:44
↓ 760 callers
Class
Field
litellm-rust/crates/python-bridge/src/coercion.rs:82
↓ 723 callers
Class
GenericLiteLLMParams
LiteLLM Params without 'model' arg (used across completion / assistants api)
litellm/types/router.py:583
↓ 709 callers
Class
MCPServerManager
litellm/proxy/mcp_server/mcp_server_manager.py:1909
↓ 507 callers
Class
LiteLLM_UserTable
litellm/models/user.py:19
↓ 487 callers
Class
AsyncHTTPHandler
litellm/llms/custom_httpx/http_handler.py:717
↓ 465 callers
Class
Request
tests/unit/proxy/auth/test_user_api_key_auth.py:29
↓ 460 callers
Class
LiteLLM_TeamTable
litellm/models/team.py:86
↓ 433 callers
Class
ProxyLogging
Logging/Custom Handlers for proxy. Implemented mainly to: - log successful/failed db read/writes - support the max parallel request
litellm/proxy/utils.py:1958
↓ 430 callers
Class
HTTPHandler
litellm/llms/custom_httpx/http_handler.py:1565
↓ 430 callers
Class
ProxyConfig
Abstraction class on top of config loading/updating logic. Gives us one place to control all config updating logic.
litellm/proxy/proxy_server.py:5930
↓ 425 callers
Class
RoutingContext
What a routing plugin (strategy, pre-call check, sub-router) may use of the router. Plugins get one instead of the Router itself.
litellm/router_api.py:1530
↓ 419 callers
Class
Choices
litellm/types/utils.py:1795
↓ 416 callers
Class
Delta
litellm/types/utils.py:1658
↓ 396 callers
Class
Number
litellm-rust/crates/cost/src/money.rs:94
↓ 392 callers
Class
Message
litellm/types/utils.py:1537
↓ 373 callers
Class
Response
tests/unit/router_utils/test_add_retry_fallback_headers.py:232
↓ 371 callers
Class
ModelResponseStream
litellm/types/utils.py:2310
↓ 369 callers
Class
UserApiKeyCache
DualCache for UserAPIKeyAuth-like payloads. Stores a Redis-safe JSON payload in memory and, with ``enable_redis_auth_cache``, in Redis, to a
litellm/proxy/common_utils/user_api_key_cache.py:50
↓ 338 callers
Class
StreamingChoices
litellm/types/utils.py:2198
↓ 324 callers
Class
LiteLLM_Params
LiteLLM Params with 'model' requirement - used for completions
litellm/types/router.py:762
↓ 323 callers
Class
InternalUsageCache
litellm/proxy/utils.py:496
↓ 306 callers
Class
Deployment
litellm/types/router.py:903
↓ 295 callers
Class
AnthropicConfig
Reference: https://docs.anthropic.com/claude/reference/messages_post to pass metadata to anthropic, it's {"user_id": "any-relevant-informati
litellm/llms/anthropic/chat/transformation.py:593
↓ 293 callers
Class
ProxyException
litellm/proxy/_types.py:3907
↓ 281 callers
Class
Logging
litellm/litellm_core_utils/litellm_logging.py:713
↓ 280 callers
Class
AmazonConverseConfig
Reference - https://docs.aws.amazon.com/bedrock/latest/APIReference/API_runtime_Converse.html #2 - https://docs.aws.amazon.com/bedrock/latest
litellm/llms/bedrock/chat/converse_transformation.py:222
↓ 271 callers
Class
GenericGuardrailAPIInputs
litellm/types/utils.py:5378
↓ 248 callers
Class
ComplexityRouter
Complexity router that classifies requests and routes to appropriate models. By default, handles requests in <1ms with zero external API cal
litellm/router_strategy/complexity_router/complexity_router.py:1417
↓ 246 callers
Class
ModelInfo
litellm/types/router.py:293
↓ 243 callers
Class
OpenTelemetry
litellm/integrations/opentelemetry.py:442
↓ 235 callers
Class
LiteLLM_JWTAuth
A class to define the roles and permissions for a LiteLLM Proxy w/ JWT Auth. Attributes: - admin_jwt_scope: The JWT scope required for p
litellm/proxy/_types.py:4938
↓ 235 callers
Class
ResponsesAPIResponse
litellm/types/llms/openai.py:1578
↓ 233 callers
Class
Member
tests/integration/authorization/test_team_member_permissions.py:80
↓ 232 callers
Class
LiteLLM_ObjectPermissionTable
Represents a LiteLLM_ObjectPermissionTable record
litellm/models/object_permission.py:11
↓ 218 callers
Class
ProxyBaseLLMRequestProcessing
litellm/proxy/common_request_processing.py:2343
↓ 217 callers
Class
PromptTokensDetailsWrapper
litellm/types/utils.py:1901
↓ 212 callers
Class
InMemoryCache
litellm/caching/in_memory_cache.py:31
↓ 211 callers
Class
LiteLLMParamsBody
POST /model/new litellm_params: `model` is the only required field; `api_key` et al may be an `os.environ/FOO` reference the proxy resolves at cal
tests/e2e/models.py:1354
↓ 206 callers
Class
LiteLLM_VerificationToken
litellm/models/verification_token.py:16
↓ 199 callers
Class
OpenAIResponsesHandler
Handler for processing OpenAI Responses API with guardrails. This class provides methods to: 1. Process input (pre-call hook) 2. Pro
litellm/llms/openai/responses/guardrail_translation/handler.py:767
↓ 192 callers
Class
UpdateKeyRequest
litellm/types/proxy/management_endpoints/key_requests.py:104
↓ 192 callers
Class
_PROXY_MaxParallelRequestsHandler
litellm/proxy/hooks/parallel_request_limiter.py:78
↓ 191 callers
Class
CustomStreamWrapper
litellm/litellm_core_utils/streaming_handler.py:409
↓ 183 callers
Class
ChangeSet
Runtime changes for ``apply``. Applied in this order: ``assign``, ``settings``, ``upsert``, ``delete``, ``redis_cache``. - ``assign``: r
litellm/router_api.py:1420
↓ 182 callers
Class
LiteLLMAnthropicMessagesAdapter
litellm/llms/anthropic/messages/adapters/transformation.py:485
↓ 177 callers
Class
ImageResponse
litellm/types/utils.py:2847
↓ 175 callers
Class
GenerateKeyRequest
litellm/types/proxy/management_endpoints/key_requests.py:75
↓ 171 callers
Class
VertexGeminiConfig
Reference: https://cloud.google.com/vertex-ai/docs/generative-ai/chat/test-chat-prompts Reference: https://cloud.google.com/vertex-ai/generat
litellm/llms/vertex_ai/gemini/vertex_and_google_ai_studio_gemini.py:734
↓ 168 callers
Class
OpenTelemetryV2Config
litellm/integrations/otel/model/config.py:148
↓ 163 callers
Class
MCPTool
litellm/types/mcp_server/tool_registry.py:9
↓ 160 callers
Class
DBSpendUpdateWriter
Module responsible for 1. Writing spend increments to either in memory list of transactions or to redis 2. Reading increments from redis
litellm/proxy/db/db_spend_update_writer.py:675
↓ 160 callers
Class
Function
litellm/types/utils.py:1330
↓ 159 callers
Class
JWTHandler
- treat the sub id passed in as the user id - return an error if id making request doesn't exist in proxy user table - track spend agains
litellm/proxy/auth/handle_jwt.py:233
↓ 156 callers
Class
BaseLLMHTTPHandler
litellm/llms/custom_httpx/llm_http_handler.py:711
↓ 151 callers
Class
BedrockGuardrail
litellm/proxy/guardrails/guardrail_hooks/bedrock_guardrails.py:276
↓ 150 callers
Class
NotImplementedError
litellm/proxy/swagger/swagger-ui-bundle.js:2
↓ 148 callers
Class
BaseAWSLLM
litellm/llms/bedrock/base_aws_llm.py:157
↓ 148 callers
Class
EmbeddingResponse
litellm/types/utils.py:2515
↓ 147 callers
Class
Cache
litellm/caching/caching.py:128
↓ 147 callers
Class
LitellmParams
litellm/types/guardrails.py:1401
↓ 147 callers
Class
updateDeployment
litellm/types/router.py:801
↓ 145 callers
Class
LiteLLM_BudgetTable
Represents user-controllable params for a LiteLLM_BudgetTable record. Budget-write paths use `model_fields.keys()` on this class as an allowlist
litellm/models/budget.py:16
↓ 140 callers
Class
Object
litellm-rust/crates/config/src/value.rs:10
↓ 138 callers
Class
ChatMessage
tests/e2e/models.py:232
↓ 134 callers
Class
S3Logger
litellm/integrations/s3_v2.py:157
↓ 132 callers
Class
RealTimeStreaming
litellm/litellm_core_utils/realtime_streaming.py:485
↓ 131 callers
Class
PrometheusLogger
litellm/integrations/prometheus.py:493
↓ 130 callers
Class
CredentialItem
litellm/models/credentials.py:25
↓ 125 callers
Class
Headers
Base for header models. Subclasses may alias to hyphenated header names (e.g. ``x-litellm-api-key``); serialization uses by_alias.
tests/e2e/e2e_http.py:33
↓ 124 callers
Class
AnthropicMessagesHandler
Process Anthropic messages with guardrails. In-sequence system entries are untrusted client input. This handler scans and preserves them thro
litellm/llms/anthropic/chat/guardrail_translation/handler.py:577
↓ 121 callers
Class
ContentFilterGuardrail
Content filter guardrail that detects sensitive information using: - Prebuilt regex patterns (SSN, credit cards, API keys, etc.) - Custom
litellm/proxy/guardrails/guardrail_hooks/litellm_content_filter/content_filter.py:341
↓ 118 callers
Class
CustomGuardrail
litellm/integrations/custom_guardrail.py:389
↓ 117 callers
Class
LiteLLM_TeamTableCachedObj
litellm/models/team.py:136
↓ 116 callers
Class
AmazonAnthropicClaudeMessagesConfig
Call Claude model family in the /v1/messages API spec Supports anthropic_beta parameter for beta features.
litellm/llms/bedrock/messages/invoke_transformations/anthropic_claude_transformation.py:89
↓ 115 callers
Class
LiteLLMResponsesTransformationHandler
Handler for transforming /chat/completions api requests to litellm.responses requests
litellm/completion_extras/litellm_responses_transformation/transformation.py:883
↓ 112 callers
Class
UpdateTeamRequest
UpdateTeamRequest, used by /team/update when you need to update a team team_id: str team_alias: Optional[str] = None organization_id
litellm/proxy/_types.py:1985
↓ 112 callers
Class
WebSearchInterceptionLogger
CustomLogger that intercepts WebSearch tool calls for models that don't natively support web search. Implements agentic loop: 1. Det
litellm/integrations/websearch_interception/handler.py:266
↓ 110 callers
Class
UnifiedLLMGuardrails
litellm/proxy/guardrails/guardrail_hooks/unified_guardrail/unified_guardrail.py:247
↓ 108 callers
Class
ChatBody
tests/e2e/models.py:350
↓ 108 callers
Class
TextContent
litellm/types/interactions/generated.py:378
↓ 106 callers
Class
ChatCompletionMessageToolCall
litellm/types/utils.py:1435
↓ 105 callers
Class
VertexBase
litellm/llms/vertex_ai/vertex_llm_base.py:105
↓ 101 callers
Class
AgentResponse
litellm/types/agents.py:314
↓ 101 callers
Class
BedrockFilesConfig
Config for Bedrock Files - handles S3 uploads for Bedrock batch processing
litellm/llms/bedrock/files/transformation.py:590
↓ 99 callers
Class
LiteLLM_MCPServerTable
Represents a LiteLLM_MCPServerTable record
litellm/models/mcp_server.py:49
↓ 99 callers
Struct
ModelResponse
ModelResponse represents a response from the API containing model information.
terraform/provider/litellm/types.go:21
↓ 98 callers
Class
AnthropicStreamWrapper
- first chunk return 'message_start' - content block must be started and stopped - finish_reason must map exactly to anthropic reason, el
litellm/llms/anthropic/messages/adapters/streaming_iterator.py:623
↓ 98 callers
Class
BedrockError
litellm/llms/bedrock/common_utils.py:108
↓ 98 callers
Class
FireworksAIConfig
Reference: https://docs.fireworks.ai/api-reference/post-chatcompletions The class `FireworksAIConfig` provides configuration for the Firewor
litellm/llms/fireworks_ai/chat/transformation.py:161
↓ 97 callers
Class
BedrockRealtimeConfig
Configuration for Bedrock Nova Sonic realtime transformations.
litellm/llms/bedrock/realtime/transformation.py:211
next →
1–100 of 26,341, ranked by callers