OpenRouterLLMClient

Implementation of LLMClient for OpenRouter API. OpenRouter is an API that routes requests to multiple LLM providers.

settings

The base URL and timeouts for the OpenRouter API, defaults to "https://openrouter.ai" and 900s

httpClient

A fully configured KoogHttpClient for making API requests. Use the secondary constructor to create a Ktor-backed client configured with an API key.

clock

Clock instance used for tracking response metadata timestamps.

Constructors

constructor(settings: OpenRouterClientSettings = OpenRouterClientSettings(), httpClient: KoogHttpClient, clock: Clock = Clock.System, toolsConverter: OpenAICompatibleToolDescriptorSchemaGenerator = OpenAICompatibleToolDescriptorSchemaGenerator())

constructor(apiKey: String, settings: OpenRouterClientSettings = OpenRouterClientSettings(), baseClient: HttpClient = HttpClient(), clock: Clock = Clock.System, toolsConverter: OpenAICompatibleToolDescriptorSchemaGenerator = OpenAICompatibleToolDescriptorSchemaGenerator())

open override val clientName: String

open override fun close()

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

Embeds the given text using the OpenRouter embeddings API.

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

Batch embedding is not supported by the OpenRouter API.

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

Returns the specific implementation of the LLMProvider associated with this client.

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

Fetches the list of available models from the OpenRouter service. https://openrouter.ai/docs/api/api-reference/models/get-models

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