Google releases Gen AI SDK for Kotlin 1.0, offering multiplatform access to Gemini

SDK Release for Kotlin Developers

Google announced the 1.0 release of the Google Gen AI SDK for Kotlin. This SDK is designed to allow Kotlin developers to integrate generative AI capabilities into their applications without needing to use raw HTTP clients or bridge Java libraries. The SDK is available on GitHub under googleapis/kotlin-genai.

Multiplatform Support and Idiomatic Kotlin

The SDK is built as a Kotlin Multiplatform (KMP) library, supporting JVM (backend services, serverless functions, desktop) and Android targets. It incorporates idiomatic Kotlin features such as Coroutines, asynchronous Flow streaming, and immutable data classes. This design aims to provide a consistent development experience across different platforms.

Gemini API Integration

The SDK offers a unified interface for interacting with both the Gemini Developer API (Google AI Studio) and the Gemini Enterprise Agent Platform (on Google Cloud). Developers can configure the SDK to use either platform with minimal changes. Authentication is handled automatically based on environment variables like GEMINI_API_KEY or GOOGLE_API_KEY.

Key Functionality

The SDK provides capabilities for unary and streaming text generation, as well as multi-turn chat. The Client class manages HTTP connections and authentication. For streaming responses, it uses Kotlin Coroutine Flow, delivering token chunks in real time. A dedicated chat service within the SDK manages conversation history, appends turns, formats conversations, and handles function calling automatically.