The Google Gen AI Go SDK is Google’s official open-source client library for building applications with Gemini and other Google generative AI models from Go. Hosted on GitHub with about 1,200 stars and updated continuously through October 2026, it provides a single, idiomatic interface to both the Gemini Developer API and Vertex AI (the Gemini Enterprise Agent Platform).
Go teams building on Google’s AI stack previously had to choose between separate clients for the consumer-facing Gemini API and the enterprise Vertex AI platform. This SDK unifies them: the same code can target the Developer API with an API key or the enterprise backend with Google Cloud credentials, switched through configuration. It covers text and multimodal generation, streaming, function calling, structured output and other core Gemini capabilities, with strongly typed Go structs throughout. Because it is maintained by Google, it tracks new model releases and API features closely, and it is the recommended path for new Go projects using Gemini, superseding older generative AI clients.
The package is organised around a single client type with sub-services for models, chats, files and other API areas, and it ships with examples and tests in the repository. It is the successor to Google’s earlier generative AI Go client, which is now deprecated in its favour.
The SDK is free and open source under the Apache-2.0 license. Costs come only from your own Gemini API or Google Cloud usage, billed by Google under your account; the library itself carries no fee.
Go developers and platform teams who want the official, supported route to Gemini models, whether prototyping against the Developer API or deploying on Google Cloud’s enterprise platform.