litellm の API リファレンス

litellm (BerriAI/litellm) の公開 API 400 件 —— クラス 181、関数 132、メソッド 87。実際のソースを静的解析して抽出した正確なシグネチャを掲載しています。

種別	件数
クラス	181
関数	132
メソッド	87

API 一覧

func.github.workflows.run_llm_translation_tests.format_duration(seconds:float) -> str

Format duration in human-readable format

func.github.workflows.run_llm_translation_tests.get_provider_from_test_file(test_file:str) -> str

Map test file names to provider names

func.github.workflows.run_llm_translation_tests.print_colored(message:str, color:str=Colors.RESET)

Print colored message to terminal

funccookbook.anthropic_agent_sdk.agent_with_mcp.main()

Run interactive chat with MCP

classcookbook.anthropic_agent_sdk.common.Config

Configuration for LiteLLM Gateway connection

funccookbook.anthropic_agent_sdk.common.fetch_available_models(base_url:str, api_key:str) -> list[str]

Fetch available models from LiteLLM proxy /models endpoint

funccookbook.anthropic_agent_sdk.common.handle_model_list(available_models:list[str], current_model:str)

Display available models

funccookbook.anthropic_agent_sdk.common.print_header(base_url:str, current_model:str, has_mcp:bool=False)

Print the chat header

funccookbook.anthropic_agent_sdk.common.stream_response(client, user_input:str)

Stream response from the agent

funccookbook.anthropic_agent_sdk.main.interactive_chat()

Interactive CLI chat with the agent

funccookbook.anthropic_agent_sdk.main.main()

Run interactive chat

funccookbook.litellm_proxy_server.braintrust_prompt_wrapper_server.health_check()

Health check endpoint.

funccookbook.litellm_proxy_server.braintrust_prompt_wrapper_server.main()

Run the server.

funccookbook.litellm_proxy_server.cli_token_usage.main()

Using CLI token with LiteLLM SDK

funccookbook.livekit_agent_sdk.main.main()

Run the voice agent

funccookbook.livekit_agent_sdk.main.run_voice_agent()

Simple voice agent that: 1.

classcookbook.mock_guardrail_server.mock_bedrock_guardrail_server.GuardrailConfig

Configuration for mock guardrail behavior

funccookbook.mock_guardrail_server.mock_bedrock_guardrail_server.check_blocked_topics(text:str) -> Optional[TopicPolicy]

Check if text contains blocked topics

funccookbook.mock_guardrail_server.mock_bedrock_guardrail_server.check_blocked_words(text:str) -> Optional[WordPolicy]

Check if text contains blocked words

funccookbook.mock_guardrail_server.mock_bedrock_guardrail_server.health()

Health check endpoint

funccookbook.mock_guardrail_server.mock_bedrock_guardrail_server.root()

Health check endpoint

classcookbook.mock_prompt_management_server.mock_prompt_management_server.MessageContent

A single message in the prompt template

funccookbook.mock_prompt_management_server.mock_prompt_management_server.health_check()

Health check endpoint

func

cookbook.mock_prompt_management_server.mock_prompt_management_server.list_prompts(authorization:Optional[str]=Header(None))

List all available prompts.

func

cookbook.mock_prompt_management_server.mock_prompt_management_server.verify_api_key(authorization:Optional[str]=Header(None)) -> bool

Verify the API key from the Authorization header.

classcookbook.nova_sonic_realtime.RealtimeClient

Client for LiteLLM realtime API with audio support.

methodcookbook.nova_sonic_realtime.RealtimeClient.close()

Close the connection and cleanup.

methodcookbook.nova_sonic_realtime.RealtimeClient.send_audio_chunk(audio_bytes:bytes)

Send audio chunk to server.

methodcookbook.nova_sonic_realtime.RealtimeClient.send_session_update()

Send session configuration.

funccookbook.nova_sonic_realtime.main()

Main function to run the realtime client.

classcookbook.veo_video_generation.VeoVideoGenerator

Complete Veo video generation client using LiteLLM proxy.

method

cookbook.veo_video_generation.VeoVideoGenerator.download_video(video_uri:str, output_filename:str='generated_video.mp4') -> bool

Download the generated video file.

methodcookbook.veo_video_generation.VeoVideoGenerator.generate_and_download(prompt:str, output_filename:str=None) -> bool

Complete workflow: generate video and download it.

methodcookbook.veo_video_generation.VeoVideoGenerator.generate_video(prompt:str) -> Optional[str]

Initiate video generation with Veo.

funccookbook.veo_video_generation.main()

Example usage of the VeoVideoGenerator.

funcdb_scripts.migrate_keys.migrate_verification_tokens()

Main migration function

funcdb_scripts.migrate_keys.parse_csv_value(value:str, field_type:str) -> Any

Parse CSV values according to their expected types

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.adafruit.AdafruitKeyDetector

Scans for Adafruit keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.adobe.AdobeSecretDetector

Scans for Adobe client keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.age_secret_key.AgeSecretKeyDetector

Scans for Age secret keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.airtable_api_key.AirtableApiKeyDetector

Scans for Airtable API keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.algolia_api_key.AlgoliaApiKeyDetector

Scans for Algolia API keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.atlassian_api_token.AtlassianApiTokenDetector

Scans for Atlassian API tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.beamer_api_token.BeamerApiTokenDetector

Scans for Beamer API tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.clojars_api_token.ClojarsApiTokenDetector

Scans for Clojars API tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.codecov_access_token.CodecovAccessTokenDetector

Scans for Codecov Access Token.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.coinbase_access_token.CoinbaseAccessTokenDetector

Scans for Coinbase Access Token.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.databricks_api_token.DatabricksApiTokenDetector

Scans for Databricks API token.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.datadog_access_token.DatadogAccessTokenDetector

Scans for Datadog Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.digitalocean.DigitaloceanDetector

Scans for various DigitalOcean Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.discord.DiscordDetector

Scans for various Discord Client Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.doppler_api_token.DopplerApiTokenDetector

Scans for Doppler API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.droneci_access_token.DroneciAccessTokenDetector

Scans for Droneci Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.dropbox.DropboxDetector

Scans for various Dropbox Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.duffel_api_token.DuffelApiTokenDetector

Scans for Duffel API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.dynatrace_api_token.DynatraceApiTokenDetector

Scans for Dynatrace API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.easypost.EasyPostDetector

Scans for various EasyPost Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.etsy_access_token.EtsyAccessTokenDetector

Scans for Etsy Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.fastly_api_token.FastlyApiKeyDetector

Scans for Fastly API keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.finnhub_access_token.FinnhubAccessTokenDetector

Scans for Finnhub Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.flickr_access_token.FlickrAccessTokenDetector

Scans for Flickr Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.flutterwave.FlutterwaveDetector

Scans for Flutterwave API Keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.frameio_api_token.FrameIoApiTokenDetector

Scans for Frame.io API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.gcp_api_key.GCPApiKeyDetector

Scans for GCP API keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.github_token.GitHubTokenCustomDetector

Scans for GitHub tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.gitlab.GitLabDetector

Scans for GitLab Secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.gitter_access_token.GitterAccessTokenDetector

Scans for Gitter Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.gocardless_api_token.GoCardlessApiTokenDetector

Scans for GoCardless API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.grafana.GrafanaDetector

Scans for Grafana Secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.heroku_api_key.HerokuApiKeyDetector

Scans for Heroku API Keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.hubspot_api_key.HubSpotApiTokenDetector

Scans for HubSpot API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.huggingface.HuggingFaceDetector

Scans for Hugging Face Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.intercom_api_key.IntercomApiTokenDetector

Scans for Intercom API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.jfrog.JFrogDetector

Scans for JFrog-related secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.jwt.JWTBase64Detector

Scans for Base64-encoded JSON Web Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.kraken_access_token.KrakenAccessTokenDetector

Scans for Kraken Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.linear.LinearDetector

Scans for Linear secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.linkedin.LinkedInDetector

Scans for LinkedIn secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.lob.LobDetector

Scans for Lob secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.mailgun.MailgunDetector

Scans for Mailgun secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.mapbox_api_token.MapBoxApiTokenDetector

Scans for MapBox API tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.messagebird.MessageBirdDetector

Scans for MessageBird secrets.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.netlify_access_token.NetlifyAccessTokenDetector

Scans for Netlify Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.new_relic.NewRelicDetector

Scans for New Relic API tokens and keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.okta_access_token.OktaAccessTokenDetector

Scans for Okta Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.openai_api_key.OpenAIApiKeyDetector

Scans for OpenAI API Keys.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.planetscale.PlanetScaleDetector

Scans for PlanetScale API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.postman_api_token.PostmanApiTokenDetector

Scans for Postman API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.prefect_api_token.PrefectApiTokenDetector

Scans for Prefect API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.pulumi_api_token.PulumiApiTokenDetector

Scans for Pulumi API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.pypi_upload_token.PyPiUploadTokenDetector

Scans for PyPI Upload Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.readme_api_token.ReadmeApiTokenDetector

Scans for Readme API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.rubygems_api_token.RubygemsApiTokenDetector

Scans for Rubygem API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.scalingo_api_token.ScalingoApiTokenDetector

Scans for Scalingo API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.sendbird.SendbirdDetector

Scans for Sendbird Access IDs and Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.sendgrid_api_token.SendGridApiTokenDetector

Scans for SendGrid API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.sendinblue_api_token.SendinBlueApiTokenDetector

Scans for SendinBlue API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.sentry_access_token.SentryAccessTokenDetector

Scans for Sentry Access Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.shippo_api_token.ShippoApiTokenDetector

Scans for Shippo API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.slack.SlackDetector

Scans for Slack tokens and webhooks.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.snyk_api_token.SnykApiTokenDetector

Scans for Snyk API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.twitch_api_token.TwitchApiTokenDetector

Scans for Twitch API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.typeform_api_token.TypeformApiTokenDetector

Scans for Typeform API Tokens.

classenterprise.litellm_enterprise.enterprise_callbacks.secrets_plugins.zendesk_secret_key.ZendeskSecretKeyDetector

Scans for Zendesk Secret Keys.

classenterprise.litellm_enterprise.enterprise_callbacks.send_emails.resend_email.ResendEmailLogger

Send emails using Resend's API.

classenterprise.litellm_enterprise.types.enterprise_callbacks.send_emails.DefaultEmailSettings

Default settings for email events

classenterprise.litellm_enterprise.types.proxy.audit_logging_endpoints.AuditLogResponse

Response model for a single audit log entry

classenterprise.litellm_enterprise.types.proxy.audit_logging_endpoints.PaginatedAuditLogResponse

Response model for paginated audit logs

classlitellm-proxy-extras.litellm_proxy_extras.prisma_toolchain.ToolchainBootstrap

Outcome of preparing the Prisma toolchain.

funclitellm._lazy_imports.get_litellm_globals() -> dict

Get the globals dictionary of the litellm module.

classlitellm._logging.SecretRedactionFilter

Scrubs known secret/credential patterns from log records.

funclitellm._uuid.uuid4()

Return a UUID4 using the selected backend.

funclitellm.a2a_protocol.card_resolver.get_agent_card_url(agent_card:'AgentCard') -> str | None

Return the agent endpoint URL from the resolved SDK card.

funclitellm.a2a_protocol.card_resolver.is_localhost_or_internal_url(url:str | None) -> bool

Check if a URL is a localhost or internal URL.

funclitellm.a2a_protocol.card_resolver.set_agent_card_url(agent_card:'AgentCard', url:str) -> None

Set the agent endpoint URL on the resolved SDK card.

classlitellm.a2a_protocol.client.A2AClient

LiteLLM wrapper for A2A agent invocation.

methodlitellm.a2a_protocol.client.A2AClient.get_agent_card() -> 'AgentCard'

Fetch the agent card from the server.

methodlitellm.a2a_protocol.client.A2AClient.send_message(request:'SendMessageRequest') -> LiteLLMSendMessageResponse

Send a message to the A2A agent.

method

litellm.a2a_protocol.client.A2AClient.send_message_streaming(request:'SendStreamingMessageRequest') -> AsyncIterator['SendStreamingMessageResponse']

Send a streaming message to the A2A agent.

classlitellm.a2a_protocol.exceptions.A2AAgentCardError

Raised when there's an issue with the agent card.

classlitellm.a2a_protocol.exceptions.A2AConnectionError

Raised when connection to an A2A agent fails.

classlitellm.a2a_protocol.exceptions.A2AError

Base exception for A2A protocol errors.

classlitellm.a2a_protocol.litellm_completion_bridge.transformation.A2AStreamingContext

Context holder for A2A streaming state.

func

litellm.a2a_protocol.main.aget_agent_card(base_url:str, timeout:float=DEFAULT_A2A_AGENT_TIMEOUT, extra_headers:dict[str, str] | None=None) -> 'AgentCard'

Fetch the agent card from an A2A agent.

classlitellm.a2a_protocol.providers.base.BaseA2AProviderConfig

Base configuration class for A2A protocol providers.

method

litellm.a2a_protocol.providers.base.BaseA2AProviderConfig.handle_non_streaming(request_id:str, params:dict[str, Any], api_base:str | None=None, **kwargs) -> dict[str, Any]

Handle non-streaming A2A request.

method

litellm.a2a_protocol.providers.base.BaseA2AProviderConfig.handle_streaming(request_id:str, params:dict[str, Any], api_base:str | None=None, **kwargs) -> AsyncIterator[dict[str, Any]]

Handle streaming A2A request.

classlitellm.a2a_protocol.providers.bedrock_agentcore.handler.BedrockAgentCoreA2AHandler

Handler for Bedrock AgentCore A2A requests.

classlitellm.a2a_protocol.providers.config_manager.A2AProviderConfigManager

Manager for A2A provider configurations.

classlitellm.a2a_protocol.providers.pydantic_ai_agents.config.PydanticAIProviderConfig

Provider configuration for Pydantic AI agents.

classlitellm.a2a_protocol.providers.pydantic_ai_agents.handler.PydanticAIHandler

Handler for Pydantic AI agent requests.

classlitellm.a2a_protocol.utils.A2ARequestUtils

Utility class for A2A request/response processing.

methodlitellm.a2a_protocol.utils.A2ARequestUtils.count_tokens(text:str) -> int

Count tokens in text using litellm.token_counter.

methodlitellm.a2a_protocol.utils.A2ARequestUtils.extract_text_from_message(message:Any) -> str

Extract text content from A2A message parts.

methodlitellm.a2a_protocol.utils.A2ARequestUtils.extract_text_from_response(response_dict:dict[str, Any]) -> str

Extract text content from A2A response result.

method

litellm.a2a_protocol.utils.A2ARequestUtils.get_input_message_from_request(request:'SendMessageRequest | SendStreamingMessageRequest') -> Any

Extract the input message from an A2A request.

funclitellm.anthropic_beta_headers_manager.get_beta_headers_config(url:str) -> dict

Public entry point — returns the beta headers config dict.

funclitellm.anthropic_beta_headers_manager.get_provider_name(provider:str) -> str

Resolve provider aliases to canonical provider names.

classlitellm.anthropic_interface.exceptions.exceptions.AnthropicErrorDetail

Inner error detail in Anthropic format.

classlitellm.anthropic_interface.exceptions.exceptions.AnthropicErrorResponse

Anthropic-formatted error response.

func

litellm.assistants.main.get_thread(custom_llm_provider:Literal['openai', 'azure'], thread_id:str, client=None, **kwargs) -> Thread

Get the thread object, given a thread_id

methodlitellm.caching.caching.Cache.add_cache(result, **kwargs)

Adds a result to the cache.

methodlitellm.caching.caching.Cache.async_add_cache(result, dynamic_cache_object:BaseCache | None=None, **kwargs)

Async implementation of add_cache

methodlitellm.caching.caching.Cache.async_get_cache(dynamic_cache_object:BaseCache | None=None, **kwargs)

Async get cache implementation.

methodlitellm.caching.caching.Cache.get_cache_key(**kwargs) -> str

Get the cache key for the given arguments.

funclitellm.caching.caching.disable_cache()

Disable the cache used by LiteLLM.

classlitellm.caching.caching_handler.CachingHandlerResponse

This is the response object for the caching handler.

methodlitellm.caching.in_memory_cache.InMemoryCache.allow_ttl_override(key:str) -> bool

Check if ttl is set for a key

methodlitellm.caching.in_memory_cache.InMemoryCache.async_get_oldest_n_keys(n:int) -> list[str]

Get the oldest n keys in the cache

methodlitellm.caching.in_memory_cache.InMemoryCache.async_get_ttl(key:str) -> int | None

Get the remaining TTL of a key in in-memory cache

methodlitellm.caching.in_memory_cache.InMemoryCache.async_set_cache_sadd(key, value:list, ttl:float | None)

Add value to set

methodlitellm.caching.in_memory_cache.InMemoryCache.evict_cache()

Eviction policy: 1.

methodlitellm.caching.redis_cache.RedisCache.async_register_script(script:str) -> Callable[..., Awaitable[Any]]

methodlitellm.caching.redis_cache.RedisCache.check_and_fix_namespace(key:str) -> str

Make sure each key starts with the given namespace

methodlitellm.caching.redis_cache.RedisCache.sync_ping() -> bool

Tests if the sync redis client is correctly setup.

classlitellm.caching.redis_cache.RedisCircuitBreaker

Tracks Redis health for a RedisCache instance.

methodlitellm.caching.redis_cache.RedisCircuitBreaker.is_open() -> bool

Returns True if Redis calls should be skipped.

classlitellm.caching.redis_semantic_cache.RedisSemanticCache

Redis-backed semantic cache for LLM responses.

methodlitellm.caching.redis_semantic_cache.RedisSemanticCache.async_set_cache(key:str, value:object, **kwargs) -> None

Asynchronously store a value in the semantic cache.

methodlitellm.caching.redis_semantic_cache.RedisSemanticCache.get_cache(key:str, **kwargs) -> object

Retrieve a semantically similar cached response.

methodlitellm.caching.redis_semantic_cache.RedisSemanticCache.set_cache(key:str, value:object, **kwargs) -> None

Store a value in the semantic cache.

classlitellm.caching.valkey_semantic_cache.ValkeySemanticCache

Valkey-backed semantic cache for LLM responses.

funclitellm.compression.content_detection.detect_content_type(content:str) -> str

Detect whether content is code, JSON, or plain text.

funclitellm.compression.message_stubbing.extract_key(message:dict, fallback_index:int, used_keys:set[str]) -> str

Extract a human-readable key for the message.

funclitellm.compression.message_stubbing.stub_message(message:dict, key:str) -> dict

Replace message content with a compact stub.

funclitellm.containers.endpoint_factory.create_sync_endpoint_function(endpoint_config:dict) -> Callable

Create a sync SDK function from endpoint config.

func

litellm.cost_calculator.batch_cost_calculator(usage:Usage, model:str, custom_llm_provider:str | None=None, model_info:ModelInfo | None=None, data_residency:str | None=None) -> tuple[float, float]

Calculate the cost of a batch job.

classlitellm.exceptions.RateLimitError

Unified rate-limit error.

classlitellm.exceptions.RateLimitType

The dimension that was exceeded when a rate-limit error fired.

funclitellm.experimental_mcp_client.client.to_basic_auth(auth_value:str) -> str

Convert auth value to Basic Auth format.

func

litellm.experimental_mcp_client.tools.call_mcp_tool(session:ClientSession, call_tool_request_params:MCPCallToolRequestParams) -> MCPCallToolResult

Call an MCP tool.

func

litellm.experimental_mcp_client.tools.call_openai_tool(session:ClientSession, openai_tool:ChatCompletionMessageToolCall) -> MCPCallToolResult

Call an OpenAI tool using MCP client.

funclitellm.experimental_mcp_client.tools.transform_mcp_tool_to_openai_tool(mcp_tool:MCPTool) -> ChatCompletionToolParam

Convert an MCP tool to an OpenAI tool.

classlitellm.files.utils.FilesAPIUtils

Utils for files API interface on litellm

method

litellm.files.utils.FilesAPIUtils.is_batch_jsonl_file(create_file_data:CreateFileRequest, extracted_file_data:ExtractedFileData) -> bool

Check if the file is a batch jsonl file

classlitellm.integrations.SlackAlerting.budget_alert_types.BaseBudgetAlertType

Base class for different budget alert types

methodlitellm.integrations.SlackAlerting.budget_alert_types.BaseBudgetAlertType.get_id(user_info:CallInfo) -> str

Return the ID to use for caching/tracking this alert

classlitellm.integrations.SlackAlerting.slack_alerting.SlackAlerting

Class for sending Slack Alerts

methodlitellm.integrations.SlackAlerting.slack_alerting.SlackAlerting.send_webhook_alert(webhook_event:WebhookEvent) -> bool

Sends structured alert to webhook, if set.

classlitellm.integrations._types.open_inference.AudioAttributes

Attributes for audio

classlitellm.integrations._types.open_inference.DocumentAttributes

Attributes for a document.

classlitellm.integrations._types.open_inference.EmbeddingAttributes

Attributes for an embedding

classlitellm.integrations._types.open_inference.ErrorAttributes

Attributes for error information in spans.

classlitellm.integrations._types.open_inference.ImageAttributes

Attributes for images

classlitellm.integrations._types.open_inference.MessageAttributes

Attributes for a message sent to or from an LLM

classlitellm.integrations._types.open_inference.RerankerAttributes

Attributes for a reranker

classlitellm.integrations._types.open_inference.ToolAttributes

Attributes for a tools

classlitellm.integrations._types.open_inference.ToolCallAttributes

Attributes for a tool call

classlitellm.integrations.arize.arize.ArizeLogger

Arize logger that sends traces to an Arize endpoint.

methodlitellm.integrations.arize.arize.ArizeLogger.get_arize_config() -> ArizeConfig

Helper function to get Arize configuration.

func

litellm.integrations.arize.prompt_initializer(litellm_params:'PromptLiteLLMParams', prompt_spec:'PromptSpec') -> 'CustomPromptManagement'

Initialize a prompt from Arize Phoenix.

funclitellm.integrations.bitbucket.set_global_bitbucket_config(config:dict) -> None

Set the global BitBucket configuration for prompt management.

classlitellm.integrations.custom_secret_manager.CustomSecretManager

Base class for implementing custom secret managers.

method

litellm.integrations.custom_secret_manager.CustomSecretManager.async_health_check(timeout:float | httpx.Timeout | None=None) -> bool

Perform a health check on your secret manager.

classlitellm.integrations.deepeval.deepeval.DeepEvalLogger

Logs litellm traces to DeepEval's platform.

func

litellm.integrations.dotprompt.prompt_initializer(litellm_params:'PromptLiteLLMParams', prompt_spec:'PromptSpec') -> 'CustomPromptManagement'

Initialize a prompt from a .prompt file.

method

litellm.integrations.dotprompt.prompt_manager.PromptManager.add_prompt(prompt_id:str, content:str, metadata:dict[str, Any] | None=None) -> None

Add a prompt template programmatically.

methodlitellm.integrations.dotprompt.prompt_manager.PromptManager.get_prompt_metadata(prompt_id:str) -> dict[str, Any] | None

Get metadata for a specific prompt.

methodlitellm.integrations.dotprompt.prompt_manager.PromptManager.json_to_prompt_file(prompt_data:dict[str, Any]) -> str

Convert JSON prompt data to .prompt file format.

methodlitellm.integrations.dotprompt.prompt_manager.PromptManager.list_prompts() -> list[str]

Get a list of all available prompt IDs.

methodlitellm.integrations.dotprompt.prompt_manager.PromptManager.prompt_file_to_json(file_path:str | Path) -> dict[str, Any]

Convert a .prompt file to JSON format.

funclitellm.integrations.dotprompt.set_global_prompt_directory(directory:str) -> None

Set the global prompt directory for dotprompt files.

classlitellm.integrations.focus.serializers.base.FocusSerializer

Base serializer turning Focus frames into bytes.

classlitellm.integrations.focus.serializers.csv.FocusCsvSerializer

Serialize normalized Focus frames to CSV bytes.

methodlitellm.integrations.focus.serializers.csv.FocusCsvSerializer.serialize(frame:pl.DataFrame) -> bytes

Encode the provided frame as a CSV payload.

methodlitellm.integrations.focus.serializers.parquet.FocusParquetSerializer.serialize(frame:pl.DataFrame) -> bytes

Encode the provided frame as a parquet payload.

funclitellm.integrations.generic_prompt_management.set_global_generic_prompt_config(config:dict) -> None

Set the global generic prompt configuration.

classlitellm.integrations.gitlab.gitlab_prompt_manager.GitLabPromptManager

GitLab prompt manager with folder support.

methodlitellm.integrations.gitlab.gitlab_prompt_manager.GitLabPromptManager.get_available_prompts() -> list[str]

Return prompt IDs.

funclitellm.integrations.gitlab.gitlab_prompt_manager.decode_prompt_id(encoded_id:str) -> str

Convert 'gitlab::invoice::extract' → 'invoice/extract'

func

litellm.integrations.gitlab.prompt_initializer(litellm_params:'PromptLiteLLMParams', prompt_spec:'PromptSpec') -> 'CustomPromptManagement'

Initialize a prompt from a Gitlab repository.

funclitellm.integrations.gitlab.set_global_gitlab_config(config:dict) -> None

Set the global gitlab configuration for prompt management.

classlitellm.integrations.levo.levo.LevoConfig

Configuration for Levo OTLP integration.

classlitellm.integrations.mock_client_factory.MockClientConfig

Configuration for creating a mock client.

methodlitellm.integrations.mock_client_factory.MockResponse.content() -> bytes

Return response content.

methodlitellm.integrations.mock_client_factory.MockResponse.json() -> dict

Return JSON response data.

methodlitellm.integrations.mock_client_factory.MockResponse.raise_for_status()

Raise exception for error status codes.

methodlitellm.integrations.mock_client_factory.MockResponse.read() -> bytes

Read response content.

methodlitellm.integrations.mock_client_factory.MockResponse.text() -> str

Return response text.

funclitellm.integrations.mock_client_factory.should_use_mock() -> bool

Determine if mock mode should be enabled.

classlitellm.integrations.opik.opik.OpikLogger

Opik Logger for logging events to an Opik Server

func

litellm.integrations.opik.opik_payload_builder.extractors.extract_tags(opik_metadata:dict[str, Any], custom_llm_provider:str | None) -> list[str]

Extract and build list of tags.

classlitellm.integrations.opik.opik_payload_builder.types.SpanPayload

Opik span payload structure

classlitellm.integrations.opik.opik_payload_builder.types.TracePayload

Opik trace payload structure

funclitellm.integrations.opik.utils.create_uuid7() -> str

Generate an RFC 9562 conformant UUIDv7 string.

func

litellm.integrations.opik.utils.get_traces_and_spans_from_payload(payload:list[dict[str, Any]]) -> tuple[list[dict[str, Any]], list[dict[str, Any]]]

Separate traces and spans from payload.

classlitellm.integrations.otel.logger.OpenTelemetryV2

The ``CustomLogger`` for OpenTelemetry.

funclitellm.integrations.otel.logger.emit_guardrail_span(entry:'StandardLoggingGuardrailInformation') -> None

Emit a guardrail span on the registered v2 OTel logger.

funclitellm.integrations.otel.mappers.resolve_mappers(names:Iterable[str]) -> list[AttributeMapper]

Resolve mapper names to instances.

funclitellm.integrations.otel.mappers.utils.drop_none(values:Mapping[str, AttrValue | None]) -> AttributeMap

Return ``values`` with ``None``-valued entries removed.

funclitellm.integrations.otel.mappers.utils.message_content(message:object) -> str | None

Extract the textual ``content`` from a chat message dict.

funclitellm.integrations.otel.mappers.utils.output_messages(data:LLMCallSpanData) -> list

The ``message`` payload of each response choice.

classlitellm.integrations.otel.mappers.weave.WeaveMapper

Maps ``LLMCallSpanData`` to Weave's vendor attributes.

classlitellm.integrations.otel.model.config.ExporterOwner

The preset that contributed an exporter.

classlitellm.integrations.otel.model.config.ExporterSpec

One span-export destination.

classlitellm.integrations.otel.model.semconv.DB

Database / cache client-span keys (OTel ``db.*`` semconv).

classlitellm.integrations.otel.model.semconv.GenAI

Canonical OTel GenAI attribute keys.

classlitellm.integrations.otel.model.semconv.GenAIOperation

Values for ``gen_ai.operation.name``.

classlitellm.integrations.otel.model.semconv.HTTP

HTTP server-span keys.

classlitellm.integrations.otel.model.semconv.JsonRpc

JSON-RPC keys carried on MCP spans.

classlitellm.integrations.otel.model.semconv.LiteLLM

Vendor-extension keys (no semconv equivalent).

classlitellm.integrations.otel.model.semconv.Metric

GenAI metric instrument names.

classlitellm.integrations.otel.model.semconv.NetworkTransport

Well-known values for ``network.transport``.

funclitellm.integrations.otel.model.spans.db_system(service_name:str) -> str | None

The ``db.system.name`` for a datastore service, else ``None``.

funclitellm.integrations.otel.model.spans.llm_call_span_name(data:'LLMCallSpanData') -> str

``"{operation} {model}"`` e.g.

funclitellm.integrations.otel.model.spans.mcp_list_tools_span_name(data:'MCPListToolsSpanData') -> str

``"{mcp.method.name}"`` i.e.

funclitellm.integrations.otel.model.spans.mcp_tool_call_span_name(data:'MCPToolCallSpanData') -> str

``"{mcp.method.name} {tool}"`` e.g.

funclitellm.integrations.otel.model.spans.proxy_request_span_name(data:'ProxyRequestSpanData') -> str

``"{method} {route}"`` (HTTP semconv).

funclitellm.integrations.otel.model.spans.root_roles() -> list[SpanRole]

Roles with no in-process parent.

funclitellm.integrations.otel.model.spans.service_span_name(data:'ServiceSpanData') -> str

``"{service} {call_type}"`` e.g.

funclitellm.integrations.otel.model.utils.to_ns(value:datetime | float | None) -> int | None

Coerce a datetime / epoch value to integer nanoseconds.

funclitellm.integrations.otel.mount.instrument_fastapi_app(app:Any) -> None

Attach OTel server-span instrumentation to the proxy FastAPI app.

funclitellm.integrations.otel.plumbing.context.get_baggage_attributes(context:Context | None=None) -> dict[str, str]

All Baggage entries on ``context`` as strings.

funclitellm.integrations.otel.plumbing.context.resolve_parent_context(threaded:Span | None=None) -> Context

The context a child span should parent under.

funclitellm.integrations.otel.plumbing.metrics.resolve_error_type(kwargs:Mapping[str, Any]) -> str

The ``error.type`` value for a failed request.

funclitellm.integrations.otel.plumbing.providers.parse_headers(raw:str | None) -> dict[str, str]

Parse an OTLP ``"k=v,k=v"`` header string into a dict.

funclitellm.integrations.prometheus.get_custom_labels_from_metadata(metadata:dict) -> dict[str, str]

Get custom labels from metadata

funclitellm.integrations.prometheus.get_custom_labels_from_tags(tags:Sequence[str]) -> dict[str, str]

Get custom labels from tags based on admin configuration.

classlitellm.integrations.weave.weave_otel.WeaveLLMObsOTELAttributes

Weave-specific LLM observability OTEL attributes.

classlitellm.integrations.weave.weave_otel.WeaveOtelLogger

Weave (W&B) OpenTelemetry Logger for LiteLLM.

classlitellm.interactions.agents.http_handler.AgentsHTTPHandler

HTTP handler for Agents API CRUD requests.

func

litellm.interactions.agents.main.get(name:str, custom_llm_provider:str | None=None, extra_headers:dict[str, Any] | None=None, timeout:float | httpx.Timeout | None=None, **kwargs) -> AgentCreateResponse | Coroutine[Any, Any, AgentCreateResponse]

Sync: Get a specific agent by name.

func

litellm.interactions.agents.main.list(custom_llm_provider:str | None=None, extra_headers:dict[str, Any] | None=None, timeout:float | httpx.Timeout | None=None, **kwargs) -> AgentListResponse | Coroutine[Any, Any, AgentListResponse]

Sync: List all agents on the provider side.

classlitellm.interactions.http_handler.InteractionsHTTPHandler

HTTP handler for Interactions API requests.

funclitellm.is_bedrock_pricing_only_model(key:str) -> bool

Excludes keys with the pattern 'bedrock/<region>/<model>'.

funclitellm.is_openai_finetune_model(key:str) -> bool

Excludes model cost keys with the pattern 'ft:<model>'.

funclitellm.litellm_core_utils.api_route_to_call_types.get_call_types_for_route(route:str) -> Sequence[CallTypes] | None

Get the CallTypes for a given API route.

funclitellm.litellm_core_utils.api_route_to_call_types.get_routes_for_call_type(call_type:CallTypes) -> list

Get all routes that use a specific CallType.

classlitellm.litellm_core_utils.audio_utils.utils.ProcessedAudioFile

Processed audio file data.

funclitellm.litellm_core_utils.audio_utils.utils.calculate_request_duration(file:FileTypes) -> float | None

Calculate audio duration from file content.

funclitellm.litellm_core_utils.cached_imports.clear_cached_imports() -> None

Clear all cached imports.

funclitellm.litellm_core_utils.cli_token_utils.get_cli_token_file_path() -> str

Get the path to the CLI token file

funclitellm.litellm_core_utils.cli_token_utils.get_litellm_gateway_api_key(expected_base_url:str | None=None) -> str | None

Get the stored CLI API key for use with LiteLLM SDK.

funclitellm.litellm_core_utils.cli_token_utils.load_cli_token() -> dict | None

Load CLI token data from file

funclitellm.litellm_core_utils.cloud_storage_security.is_managed_cloud_storage_uri(file_id:str) -> bool

True if file_id is a raw cloud-storage object URI (e.g.

funclitellm.litellm_core_utils.core_helpers.remove_items_at_indices(items:list[Any] | None, indices:Iterable[int]) -> None

Remove items from a list in-place by index

classlitellm.litellm_core_utils.dd_tracing.NullSpan

A no-op span implementation.

classlitellm.litellm_core_utils.dd_tracing.NullTracer

A no-op tracer implementation.

funclitellm.litellm_core_utils.dd_tracing.set_active_span_tag(tag_key:str, tag_value:str) -> bool

Best-effort helper to set a tag on the active Datadog span.

funclitellm.litellm_core_utils.dot_notation_indexing.is_nested_path(path:str) -> bool

Check if path requires nested handling.

classlitellm.litellm_core_utils.get_blog_posts.GetBlogPosts

Fetches, validates, and caches LiteLLM blog posts.

methodlitellm.litellm_core_utils.get_blog_posts.GetBlogPosts.fetch_rss_feed(url:str, timeout:int=5) -> str

Fetch RSS XML from a remote URL.

methodlitellm.litellm_core_utils.get_blog_posts.GetBlogPosts.get_blog_posts(url:str) -> list[dict[str, str]]

Return the blog posts list.

methodlitellm.litellm_core_utils.get_blog_posts.GetBlogPosts.load_local_blog_posts() -> list[dict[str, str]]

Load the bundled local backup blog posts.

methodlitellm.litellm_core_utils.get_blog_posts.GetBlogPosts.validate_blog_posts(posts:list[dict[str, str]]) -> bool

Return True if posts is a non-empty list.

funclitellm.litellm_core_utils.get_blog_posts.get_blog_posts(url:str) -> list[dict[str, str]]

Public entry point — returns the blog posts list.

funclitellm.litellm_core_utils.get_model_cost_map.get_model_cost_map(url:str) -> dict

Public entry point — returns the model cost map dict.

funclitellm.litellm_core_utils.llm_cost_calc.utils.get_billable_input_tokens(usage:Usage) -> int

Returns the number of billable input tokens.

func

litellm.litellm_core_utils.llm_response_utils.get_api_base.get_api_base(model:str, optional_params:dict | LiteLLM_Params) -> str | None

Returns the api base used for calling the model.

func

litellm.litellm_core_utils.prompt_templates.common_utils.extract_images_from_message(message:AllMessageValues) -> list[str]

Extract images from a message.

func

litellm.litellm_core_utils.prompt_templates.common_utils.get_content_from_model_response(response:ModelResponse | dict) -> str

Gets content from model response

func

litellm.litellm_core_utils.prompt_templates.common_utils.get_file_ids_from_messages(messages:list[AllMessageValues]) -> list[str]

Gets file ids from messages

funclitellm.litellm_core_utils.prompt_templates.common_utils.get_str_from_messages(messages:list[AllMessageValues]) -> str

Converts a list of messages to a string

func

litellm.litellm_core_utils.prompt_templates.common_utils.get_tool_call_names(tools:list[ChatCompletionToolParam]) -> list[str]

Get tool call names from tools

funclitellm.litellm_core_utils.prompt_templates.common_utils.is_function_call(optional_params:dict) -> bool

Checks if the optional params contain the function call

func

litellm.litellm_core_utils.prompt_templates.common_utils.migrate_file_to_image_url(message:'ChatCompletionFileObject') -> 'ChatCompletionImageObject'

Migrate file to image_url

func

litellm.litellm_core_utils.prompt_templates.common_utils.set_last_user_message(messages:list[AllMessageValues], content:str) -> list[AllMessageValues]

Set the last user message 1.

func

litellm.litellm_core_utils.prompt_templates.common_utils.strip_name_from_message(message:AllMessageValues, allowed_name_roles:list[str]=['user']) -> AllMessageValues

Removes 'name' from message

func

litellm.litellm_core_utils.prompt_templates.common_utils.strip_name_from_messages(messages:list[AllMessageValues], allowed_name_roles:list[str]=['user']) -> list[AllMessageValues]

Removes 'name' from messages

func

litellm.litellm_core_utils.prompt_templates.common_utils.strip_none_values_from_message(message:AllMessageValues) -> AllMessageValues

Strips None values from message

func

litellm.litellm_core_utils.prompt_templates.factory.anthropic_infer_file_id_content_type(file_id:str) -> Literal['document_url', 'container_upload']

Use when 'format' not provided.

funclitellm.litellm_core_utils.prompt_templates.factory.anthropic_messages_pt_xml(messages:list)

format messages for anthropic 1.

funclitellm.litellm_core_utils.prompt_templates.factory.get_all_tool_calls(messages:list) -> list

Returns extracted list of `tool_calls`.

funclitellm.litellm_core_utils.redact_messages.should_redact_message_logging(model_call_details:dict) -> bool

Determine if message logging should be redacted.

funclitellm.litellm_core_utils.safe_json_loads.safe_json_loads(data:str, default:Any=None) -> Any

Safely parse a JSON string.

funclitellm.litellm_core_utils.streaming_chunk_builder_utils.concatenate_base64_list(base64_strings:list[str]) -> str

Concatenates a list of base64-encoded strings.

classlitellm.litellm_core_utils.url_utils.SSRFError

Raised when a URL targets a blocked network.

funclitellm.litellm_core_utils.url_utils.async_safe_get(client:Any, url:str, **kwargs:Any) -> Any

Async version of safe_get.

funclitellm.litellm_core_utils.url_utils.encode_url_path_segment(value:Any, *field_name:str='path parameter') -> str

Percent-encode one user-controlled URL path segment.

classlitellm.llms.a2a.chat.streaming_iterator.A2AModelResponseIterator

Iterator for parsing A2A streaming responses.

method

litellm.llms.a2a.chat.streaming_iterator.A2AModelResponseIterator.chunk_parser(chunk:dict) -> GenericStreamingChunk | ModelResponseStream

Parse A2A streaming chunk to OpenAI format.

classlitellm.llms.a2a.chat.transformation.A2AConfig

Configuration for A2A (Agent-to-Agent) Protocol.

method

litellm.llms.a2a.chat.transformation.A2AConfig.get_error_class(error_message:str, status_code:int, headers:dict | httpx.Headers) -> BaseLLMException

Return appropriate error class for A2A errors

methodlitellm.llms.a2a.chat.transformation.A2AConfig.get_supported_openai_params(model:str) -> list[str]

Return list of supported OpenAI parameters

method

litellm.llms.a2a.chat.transformation.A2AConfig.map_openai_params(non_default_params:dict, optional_params:dict, model:str, drop_params:bool) -> dict

Map OpenAI parameters to A2A parameters.

classlitellm.llms.a2a.common_utils.A2AError

Base exception for A2A protocol errors

func

litellm.llms.a2a.common_utils.extract_text_from_a2a_message(message:dict[str, Any], depth:int=0, max_depth:int=10) -> str

Extract text content from A2A message parts.

funclitellm.llms.a2a.common_utils.extract_text_from_a2a_response(response_dict:dict[str, Any], max_depth:int=10) -> str

Extract text content from A2A response result.

funclitellm.llms.aiml.image_generation.cost_calculator.cost_calculator(model:str, image_response:Any) -> float

AI/ML flux image generation cost calculator

classlitellm.llms.anthropic.batches.handler.AnthropicBatchesHandler

Handler for Anthropic Message Batches API.

methodlitellm.llms.anthropic.chat.transformation.AnthropicConfig.get_max_tokens_for_model(model:str | None=None) -> int

Get the max output tokens for a given model.

method

litellm.llms.anthropic.chat.transformation.AnthropicConfig.get_prefix_prompt(messages:list[AllMessageValues]) -> str | None

Get the prefix prompt from the messages.

method

litellm.llms.anthropic.chat.transformation.AnthropicConfig.update_headers_with_optional_anthropic_beta(headers:dict, optional_params:dict) -> dict

Update headers with optional anthropic beta.

classlitellm.llms.anthropic.count_tokens.handler.AnthropicCountTokensHandler

Handler for Anthropic CountTokens API requests.

classlitellm.llms.anthropic.files.handler.AnthropicFilesHandler

Handles Anthropic Files API operations.

classlitellm.llms.anthropic.files.transformation.AnthropicFilesConfig

Transformation config for Anthropic Files API.

classlitellm.llms.anthropic.skills.transformation.AnthropicSkillsConfig

Anthropic-specific Skills API configuration

method

litellm.llms.anthropic.skills.transformation.AnthropicSkillsConfig.get_complete_url(api_base:str | None, endpoint:str, skill_id:str | None=None) -> str

Get complete URL for Anthropic Skills API

method

litellm.llms.anthropic.skills.transformation.AnthropicSkillsConfig.validate_environment(headers:dict, litellm_params:GenericLiteLLMParams | None) -> dict

Add Anthropic-specific headers

classlitellm.llms.azure.chat.gpt_5_transformation.AzureOpenAIGPT5Config

Azure specific handling for gpt-5 models.

funclitellm.llms.azure.common_utils.get_azure_ad_token(litellm_params:GenericLiteLLMParams) -> str | None

Get Azure AD token from various authentication methods.

method

litellm.llms.azure.containers.transformation.AzureContainerConfig.get_complete_url(api_base:str | None, litellm_params:dict) -> str

Build the Azure container endpoint URL.

classlitellm.llms.azure.image_generation.dall_e_2_transformation.AzureDallE2ImageGenerationConfig

Azure dall-e-2 image generation config

classlitellm.llms.azure.image_generation.dall_e_3_transformation.AzureDallE3ImageGenerationConfig

Azure dall-e-3 image generation config

classlitellm.llms.azure.image_generation.gpt_transformation.AzureGPTImageGenerationConfig

Azure gpt-image image generation config

classlitellm.llms.azure.videos.transformation.AzureVideoConfig

Configuration class for OpenAI video generation.

method

litellm.llms.azure.videos.transformation.AzureVideoConfig.get_complete_url(model:str, api_base:str | None, litellm_params:dict) -> str

Constructs a complete URL for the API request.

classlitellm.llms.azure_ai.agents.handler.AzureAIAgentsHandler

Handler for Azure AI Agent Service.

classlitellm.llms.azure_ai.agents.transformation.AzureAIAgentsConfig

Configuration for Azure AI Agent Service API.

methodlitellm.llms.azure_ai.agents.transformation.AzureAIAgentsConfig.get_agent_id_from_model(model:str) -> str

Extract agent ID from the model string.

methodlitellm.llms.azure_ai.agents.transformation.AzureAIAgentsConfig.is_azure_ai_agents_route(model:str) -> bool

Check if the model is an Azure AI Agents route.

classlitellm.llms.azure_ai.anthropic.handler.AzureAnthropicChatCompletion

Azure Anthropic chat completion handler.

classlitellm.llms.azure_ai.common_utils.AzureFoundryModelInfo

Model info for Azure AI / Azure Foundry models.

method

litellm.llms.azure_ai.common_utils.AzureFoundryModelInfo.get_azure_ai_route(model:str) -> Literal['agents', 'model_router', 'default']

Get the Azure AI route for the given model.

method

litellm.llms.azure_ai.common_utils.AzureFoundryModelInfo.get_models(api_key:str | None=None, api_base:str | None=None) -> list[str]

Returns a list of models supported by Azure AI.

methodlitellm.llms.azure_ai.common_utils.AzureFoundryModelInfo.strip_model_router_prefix(model:str) -> str

Strip the model_router prefix from model name.

classlitellm.llms.azure_ai.image_edit.mai_transformation.AzureFoundryMAIImageEditConfig

Azure AI Foundry MAI image editing (e.g.

funclitellm.llms.azure_ai.image_generation.cost_calculator.cost_calculator(model:str, image_response:Any) -> float

Azure AI image generation cost calculator

classlitellm.llms.azure_ai.ocr.transformation.AzureAIOCRConfig

Azure AI OCR transformation configuration.

func

litellm.llms.base_llm.base_model_iterator.convert_model_response_to_streaming(model_response:ModelResponse) -> ModelResponseStream

Convert a ModelResponse to ModelResponseStream.

methodlitellm.llms.base_llm.chat.transformation.BaseConfig.post_stream_processing(stream:Any) -> Any

Hook for providers to post-process streaming responses.

classlitellm.llms.base_llm.evals.transformation.BaseEvalsAPIConfig

Base configuration for Evals API providers

method

litellm.llms.base_llm.evals.transformation.BaseEvalsAPIConfig.get_error_class(error_message:str, status_code:int, headers:dict) -> Exception

Get appropriate error class for the provider.

classlitellm.llms.base_llm.files.azure_blob_storage_backend.AzureBlobStorageBackend

Azure Blob Storage backend implementation.

methodlitellm.llms.base_llm.files.azure_blob_storage_backend.AzureBlobStorageBackend.download_file(storage_url:str) -> bytes

Download a file from Azure Blob Storage.

classlitellm.llms.base_llm.files.storage_backend.BaseFileStorageBackend

Abstract base class for file storage backends.

methodlitellm.llms.base_llm.files.storage_backend.BaseFileStorageBackend.delete_file(storage_url:str) -> None

Delete a file from the storage backend.

methodlitellm.llms.base_llm.files.storage_backend.BaseFileStorageBackend.download_file(storage_url:str) -> bytes

Download a file from the storage backend.

funclitellm.llms.base_llm.managed_resources.utils.decode_unified_id(encoded_unified_id:str) -> str | None

Decode a base64 encoded unified ID.

funclitellm.llms.base_llm.managed_resources.utils.encode_unified_id(unified_id_string:str) -> str

Encode a unified ID string to base64.

funclitellm.llms.base_llm.managed_resources.utils.extract_model_id_from_unified_id(unified_id:str) -> str | None

Extract model ID from a unified resource ID.

funclitellm.llms.base_llm.managed_resources.utils.extract_resource_type_from_unified_id(unified_id:str) -> str | None

Extract resource type from a unified resource ID.

funclitellm.llms.base_llm.managed_resources.utils.extract_unified_uuid_from_unified_id(unified_id:str) -> str | None

Extract the UUID from a unified resource ID.

funclitellm.llms.base_llm.managed_resources.utils.parse_unified_id(unified_id:str) -> dict | None

Parse a unified ID into its components.

classlitellm.llms.base_llm.ocr.transformation.BaseOCRConfig

Base configuration for OCR transformations.

method

litellm.llms.base_llm.ocr.transformation.BaseOCRConfig.get_complete_url(api_base:str | None, model:str, optional_params:dict, litellm_params:dict | None=None, **kwargs) -> str

Get complete URL for OCR endpoint.

method

litellm.llms.base_llm.ocr.transformation.BaseOCRConfig.get_error_class(error_message:str, status_code:int, headers:dict) -> Exception

Get appropriate error class for the provider.

methodlitellm.llms.base_llm.ocr.transformation.BaseOCRConfig.get_supported_ocr_params(model:str) -> list

Get supported OCR parameters for this provider.

classlitellm.llms.base_llm.ocr.transformation.OCRPage

Single page from OCR response.

classlitellm.llms.base_llm.ocr.transformation.OCRPageDimensions

Page dimensions from OCR response.

classlitellm.llms.base_llm.ocr.transformation.OCRPageImage

Image extracted from OCR page.

classlitellm.llms.base_llm.ocr.transformation.OCRRequestData

OCR request data structure.

classlitellm.llms.base_llm.ocr.transformation.OCRResponse

Standard OCR response format.

classlitellm.llms.base_llm.ocr.transformation.OCRUsageInfo

Usage information from OCR response.

classlitellm.llms.base_llm.sandbox.transformation.BaseSandboxConfig

Provider-agnostic sandbox operations.

classlitellm.llms.base_llm.sandbox.transformation.CodeExecutionResult

Passthrough of the sandbox's own execution output.

classlitellm.llms.base_llm.sandbox.transformation.ContainerHandle

A live sandbox container.

classlitellm.llms.base_llm.search.transformation.BaseSearchConfig

Base configuration for Search transformations.

method

litellm.llms.base_llm.search.transformation.BaseSearchConfig.get_complete_url(api_base:str | None, optional_params:dict, data:dict | list[dict] | None=None, **kwargs) -> str

Get complete URL for Search endpoint.

method

litellm.llms.base_llm.search.transformation.BaseSearchConfig.get_error_class(error_message:str, status_code:int, headers:dict) -> Exception

Get appropriate error class for the provider.

methodlitellm.llms.base_llm.search.transformation.BaseSearchConfig.get_http_method() -> Literal['GET', 'POST']

Get HTTP method for search requests.

methodlitellm.llms.base_llm.search.transformation.BaseSearchConfig.ui_friendly_name() -> str

UI-friendly name for the search provider.

method

litellm.llms.base_llm.search.transformation.BaseSearchConfig.validate_environment(headers:dict, api_key:str | None=None, api_base:str | None=None, **kwargs) -> dict

Validate environment and return headers.

classlitellm.llms.base_llm.search.transformation.SearchResponse

Standard Search response format.

classlitellm.llms.base_llm.search.transformation.SearchResult

Single search result.

classlitellm.llms.base_llm.skills.transformation.BaseSkillsAPIConfig

Base configuration for Skills API providers

method

litellm.llms.base_llm.skills.transformation.BaseSkillsAPIConfig.get_error_class(error_message:str, status_code:int, headers:dict) -> Exception

Get appropriate error class for the provider.

classlitellm.llms.bedrock.batches.handler.BedrockBatchesHandler

Handler for Bedrock Batches.

classlitellm.llms.bedrock.common_utils.BedrockEventStreamDecoderBase

Base class for event stream decoding for Bedrock

method

litellm.llms.bedrock.common_utils.CommonBatchFilesUtils.extract_model_from_s3_file_path(s3_uri:str, optional_params:dict) -> str

Extract model ID from S3 file path.

methodlitellm.llms.bedrock.common_utils.CommonBatchFilesUtils.generate_unique_job_name(model:str, prefix:str='litellm') -> str

Generate a unique job name for AWS services.

method

litellm.llms.bedrock.common_utils.CommonBatchFilesUtils.get_error_class(error_message:str, status_code:int, headers:dict | httpx.Headers) -> BaseLLMException

Get Bedrock-specific error class.

methodlitellm.llms.bedrock.common_utils.CommonBatchFilesUtils.parse_s3_uri(s3_uri:str) -> tuple

Parse S3 URI into bucket and key components.

funclitellm.llms.bedrock.common_utils.extract_model_name_from_bedrock_arn(model:str) -> str

Extract the model name from an AWS Bedrock ARN.

funclitellm.llms.bedrock.common_utils.get_bedrock_base_model(model:str) -> str

Get the base model from the given model name.

funclitellm.llms.bedrock.common_utils.remove_custom_field_from_tools(request_body:dict) -> None

Remove ``custom`` field from each tool in the request body.

funclitellm.llms.bedrock.common_utils.strip_bedrock_routing_prefix(model:str) -> str

Strip LiteLLM routing prefixes from model name.

funclitellm.llms.bedrock.files.transformation.extract_s3_uri_from_file_id(file_id:str) -> str

Resolve a Bedrock file id to its S3 URI.

classlitellm.llms.bedrock.image_edit.handler.BedrockImageEdit

Bedrock Image Edit handler

classlitellm.llms.bedrock.image_generation.image_handler.BedrockImageGeneration

Bedrock Image Generation handler

この情報について

掲載しているシグネチャは BerriAI/litellm の公開ソースコードを Python の ast モジュールで静的解析し、引数名・デフォルト値・型注釈・戻り値型をそのまま抽出したものです。実装コードは保存していません。詳しくは仕組みの解説をご覧ください。

収録ライブラリ一覧（全 805 件）へ戻る