Back to feed
Artificial intelligencePrimary source

Google makes Gemini 4 Hydrogen available for coding and agentic workflows

Google announced that Gemini 4 Hydrogen is now available to developers. The model supports repository coding, debugging, interactive web development and multimodal workflows, with four configurable thinking levels.

Preferred on Google
Text size

Image: blog.google · source articleEditorial excerpt for news reporting

Google has made Gemini 4 Hydrogen available to developers for everyday coding and agentic workflows.

The model is designed for rapid iteration and supports repository coding and debugging, interactive web development and multimodal tasks. Gemini 4 Hydrogen offers four configurable thinking levels.

Model workflows and integrations

Hydrogen can run independently or alongside Gemini 4 Carbon and Gemini 4 Argon. Google says it follows instructions more closely and produces fewer fabricated claims, but the announcement does not provide test figures.

Developers can use the Gemini API in Google AI Studio and Android Studio. The model is also available through Gemini for business, while paid Google AI Pro and Ultra subscribers can try it in Google Antigravity.

Google’s Gemini API What’s New Guide contains model details, setup instructions and controls for thinking levels. The Google Antigravity blog describes Orchestrator, a feature that pairs Hydrogen with Gemini 4 Carbon.

What we know

  • Gemini 4 Hydrogen is available to developers for coding and agentic workflows.
  • The model supports repository coding, debugging, web development and multimodal tasks.
  • Hydrogen has four configurable thinking levels.
  • It can be accessed through Google AI Studio, Android Studio, Gemini for business and Google Antigravity for paid subscribers.

Trust: Primary source

  • The story cites a primary statement. It establishes what its author said, but does not guarantee every claim is true.
  • Sources: Google Blog.
Related news
View sources1

COMMUNITY

Discussion

—

Loading comments…