prompt-executor-google-client/ai.koog.prompt.executor.clients.google/GoogleLLMClient

GoogleLLMClient

open class GoogleLLMClient @JvmOverloads constructor(settings: GoogleClientSettings = GoogleClientSettings(), httpClient: KoogHttpClient, clock: Clock = Clock.System) : LLMClient(source)

Implementation of LLMClient for Google's Gemini API.

This client supports both standard and streaming text generation with optional tool calling capabilities.

Parameters

settings

Custom client settings, defaults to standard API endpoint and timeouts

httpClient

A preconfigured Koog HTTP client used for API calls. Must have authentication and other request defaults already embedded. To use a Ktor-backed client with standard defaults, use the secondary constructor that accepts an API key and an io.ktor.client.HttpClient.

clock

Clock instance used for tracking response metadata timestamps.

Constructors

GoogleLLMClient

@JvmOverloads

constructor(settings: GoogleClientSettings = GoogleClientSettings(), httpClient: KoogHttpClient, clock: Clock = Clock.System)

@JvmOverloads

constructor(apiKey: String, settings: GoogleClientSettings = GoogleClientSettings(), baseClient: HttpClient = HttpClient(), clock: Clock = Clock.System)

Secondary constructor for creating a GoogleLLMClient backed with a Ktor HTTP client.

Properties

clientName

open override val clientName: String

Provides the Large Language Model (LLM) provider associated with this client.

Functions

open override fun close()

embed

open suspend override fun embed(text: String, model: LLModel): List<Double>

Embeds the given text using the Google AI embeddings API.

open suspend override fun embed(inputs: List<String>, model: LLModel): List<List<Double>>

Embeds the given inputs using the Google AI batch embeddings API.

execute

open suspend override fun execute(prompt: Prompt, model: LLModel, tools: List<ToolDescriptor>): List<Message.Response>

executeMultipleChoices

open suspend override fun executeMultipleChoices(prompt: Prompt, model: LLModel, tools: List<ToolDescriptor>): List<LLMChoice>

executeStreaming

open fun executeStreaming(prompt: Prompt, model: LLModel): Flow<StreamFrame>

open override fun executeStreaming(prompt: Prompt, model: LLModel, tools: List<ToolDescriptor>): Flow<StreamFrame>

getBasicJsonSchemaGenerator

open override fun getBasicJsonSchemaGenerator(): GoogleBasicJsonSchemaGenerator

getStandardJsonSchemaGenerator

open override fun getStandardJsonSchemaGenerator(): GoogleStandardJsonSchemaGenerator

llmProvider

open override fun llmProvider(): LLMProvider

models

open suspend override fun models(): List<LLModel>

Retrieves a list of available language models supported by the Google LLM client. https://ai.google.dev/api/models#method:-models.list

moderate

open suspend override fun moderate(prompt: Prompt, model: LLModel): ModerationResult

Moderates the given prompt using the specified language model. This method is not supported by the Google API and will throw an exception when invoked.