gcp Google Cloud Blog ·

Google releases Kotlin Multiplatform SDK 1.0 for idiomatic Gemini AI access

blogaigapreviewengineerhealthcaremediagcp-pubsub
feature announcement

Google has released version 1.0 of its Gen AI SDK for Kotlin, offering idiomatic, multiplatform access to the Gemini Developer API and Enterprise Agent Platform. The SDK supports Kotlin Coroutines, Flow streaming, and immutable data classes, targeting JVM, Android, and other Kotlin Multiplatform environments. This provides developers a unified surface for text generation, multi-turn chat, multimodal analysis with Google Search Grounding, and visual generation using Gemini 3 models. It is available on GitHub and Maven Central.

  • Google Gen AI SDK for Kotlin 1.0 released for multiplatform Gemini access
  • Unary and streaming text generation with multi-turn chat support
  • Multimodal analysis with Google Search Grounding
  • Support for Gemini 3 image family for visual generation
Features (4)
  • Google Gen AI SDK for Kotlin 1.0 released for multiplatform Gemini access

    The SDK provides idiomatic Kotlin paradigms, including Coroutines and Flow streaming, for developers targeting JVM (backend, serverless, desktop) and Android. It offers a unified surface to interact with both the Gemini Developer API and the Gemini Enterprise Agent Platform.

  • Unary and streaming text generation with multi-turn chat support

    The SDK's Client class facilitates single prompt requests and low-latency streaming text generation via Kotlin Coroutine Flow. It also includes a dedicated chats service to automatically manage conversation history for multi-turn chat sessions and function calling.

  • Multimodal analysis with Google Search Grounding

    Gemini's multimodal reasoning capabilities are enhanced with external verification through Google Search Grounding. This allows for cross-checking factual claims against live web sources, useful for analyzing technical or medical diagrams.

  • Support for Gemini 3 image family for visual generation

    The SDK fully supports Google's latest image generation models, enabling developers to generate images and retrieve the output directly as Blob data in response parts. This includes capabilities for conversational editing.

Read the original announcement →

https://cloud.google.com/blog/topics/developers-practitioners/announcing-the-google-gen-ai-sdk-for-kotlin-10-idiomatic-multiplatform-access-to-gemini/

Related releases