Struct GeminiModel 

pub struct GeminiModel { /* private fields */ }

Available on crate feature gemini only.

Gemini model client wrapping the adk-gemini crate for the Llm trait.

Implementations

impl GeminiModel

pub fn new( api_key: impl Into<String>, model: impl Into<String>, ) -> Result<GeminiModel, AdkError>

Available on crate feature models only.

Create a new Gemini model client with an API key and model name.

pub fn with_retry_config(self, retry_config: RetryConfig) -> GeminiModel

Available on crate feature models only.

Set the retry configuration (builder pattern).

pub fn set_retry_config(&mut self, retry_config: RetryConfig)

Available on crate feature models only.

Set the retry configuration (mutable reference).

pub fn retry_config(&self) -> &RetryConfig

Available on crate feature models only.

Returns the current retry configuration.

pub fn with_thinking_config( self, thinking_config: ThinkingConfig, ) -> GeminiModel

Available on crate feature models only.

Set the default thinking configuration applied to every request.

Controls the model’s reasoning effort. For Gemini 3 series models, use ThinkingLevel (Low/Medium/High). For Gemini 2.5 series, use thinking_budget (token count).

Example

use adk_gemini::{ThinkingConfig, ThinkingLevel};

// Gemini 3 — level-based thinking
let model = GeminiModel::new(api_key, "gemini-3.1-pro-preview")?
    .with_thinking_config(
        ThinkingConfig::new().with_thinking_level(ThinkingLevel::Low)
    );

// Gemini 2.5 — budget-based thinking
let model = GeminiModel::new(api_key, "gemini-2.5-flash")?
    .with_thinking_config(
        ThinkingConfig::new().with_thinking_budget(2048)
    );

pub fn set_thinking_config(&mut self, thinking_config: ThinkingConfig)

Available on crate feature models only.

Set the thinking configuration (mutable reference variant).

pub fn thinking_config(&self) -> Option<&ThinkingConfig>

Available on crate feature models only.

Returns the current thinking configuration, if set.

pub async fn create_cached_content( &self, system_instruction: &str, tools: &HashMap<String, Value>, ttl_seconds: u32, ) -> Result<String, AdkError>

Available on crate feature models only.

Create a cached content resource with the given system instruction, tools, and TTL.

Returns the cache name (e.g., “cachedContents/abc123”) on success. The cache is created using the model configured on this GeminiModel instance.

pub async fn delete_cached_content(&self, name: &str) -> Result<(), AdkError>

Available on crate feature models only.

Delete a cached content resource by name.

Trait Implementations

impl CacheCapable for GeminiModel

fn create_cache<'life0, 'life1, 'life2, 'async_trait>( &'life0 self, system_instruction: &'life1 str, tools: &'life2 HashMap<String, Value>, ttl_seconds: u32, ) -> Pin<Box<dyn Future<Output = Result<String, AdkError>> + Send + 'async_trait>>

where 'life0: 'async_trait, 'life1: 'async_trait, 'life2: 'async_trait, GeminiModel: 'async_trait,

Create a cached content resource from the given system instruction, tool definitions, and TTL.

fn delete_cache<'life0, 'life1, 'async_trait>( &'life0 self, name: &'life1 str, ) -> Pin<Box<dyn Future<Output = Result<(), AdkError>> + Send + 'async_trait>>

where 'life0: 'async_trait, 'life1: 'async_trait, GeminiModel: 'async_trait,

Delete a previously created cached content resource by name.

impl Llm for GeminiModel

fn name(&self) -> &str

Returns the model identifier (e.g., “gemini-2.5-flash”).

fn schema_adapter(&self) -> &dyn SchemaAdapter

Returns the schema adapter for this provider.

fn generate_content<'life0, 'async_trait>( &'life0 self, req: LlmRequest, stream: bool, ) -> Pin<Box<dyn Future<Output = Result<Pin<Box<dyn Stream<Item = Result<LlmResponse, AdkError>> + Send>>, AdkError>> + Send + 'async_trait>>

where 'life0: 'async_trait, GeminiModel: 'async_trait,

Generates content from the given request, optionally streaming.