Module model

Available on crate feature models only.

Expand description

Model integrations (Gemini, etc.).

Provides LLM implementations:

ADK is model-agnostic - implement the Llm trait for other providers.

Available with feature: models

Modules§

anthropicanthropic: Anthropic/Claude provider implementation for ADK.
azure_aiazure-ai: Azure AI Inference provider for ADK.
bedrockbedrock: Amazon Bedrock provider implementation for ADK.
deepseekdeepseek: DeepSeek provider implementation for ADK.
geminigemini
groqgroq: Groq provider implementation for ADK.
mock
ollamaollama: Ollama local LLM provider implementation for ADK.
openaiopenai: OpenAI provider implementation for ADK.
openai_compatibleopenai: Shared OpenAI-compatible provider implementation.
openrouteropenrouter: OpenRouter provider implementation for ADK.
provider
retry
usage_tracking: Stream wrapper that records token usage on the active tracing span.

AnthropicClient: Anthropic client for Claude models.
AzureAIClient: Azure AI Inference client for models hosted on Azure AI endpoints.
AzureAIConfig: Configuration for Azure AI Inference endpoints.
AzureConfig: Configuration for Azure OpenAI Service.
AzureOpenAIClient: Azure OpenAI client.
BedrockClient: Amazon Bedrock client backed by the AWS SDK Converse API.
BedrockConfig: Configuration for Amazon Bedrock.
DeepSeekClient: DeepSeek client for deepseek-chat and deepseek-reasoner models.
DeepSeekConfig: Configuration for DeepSeek API.
GeminiModel
GroqClient: Groq client for ultra-fast LLM inference.
GroqConfig: Configuration for Groq API.
MockLlm
OllamaConfig: Configuration for connecting to an Ollama server.
OllamaModel: Ollama client for local LLM inference.
OpenAIClient: OpenAI client for standard OpenAI API and OpenAI-compatible APIs.
OpenAICompatible: Shared OpenAI-compatible client implementation.
OpenAICompatibleConfig: Configuration for OpenAI-compatible providers.
OpenAIConfig: Configuration for OpenAI API.
OpenRouterClient: Shared OpenRouter client used by the native APIs and the Llm adapter.
OpenRouterConfig: OpenRouter configuration shared by native APIs and the Llm adapter.
RetryConfig
ServerRetryHint: Hint from the server about when to retry.

ModelProvider: Canonical provider identifiers and metadata shared across ADK crates.
OpenRouterApiMode: Default API surface used by the Llm adapter.
ReasoningEffort: Reasoning effort level for OpenAI reasoning models (e.g., o1, o3).