Google releases Gemini 4 Hydrogen, built for fast coding and agent tasks

Google released Gemini 4 Hydrogen on 12 November 2026, a model it says is designed for quick execution of coding and agent tasks in everyday work. The company calls its Gemini 4 Carbon and Argon models its frontier models for deep reasoning over long problems, and says Hydrogen can be used on its own or paired with them as an efficient execution model.

Hydrogen is rolling out worldwide from launch day on every tier of Google AI Studio and the Gemini API, in the Gemini app, Gemini for business and Google Search, and in Google Antigravity for paying Google AI Ultra and Pro subscribers. Google says free Gemini app users get it through automatic model selection when it fits the prompt.

The model has a context window of 1 million tokens. Input costs $1.50 per million tokens and output $7.50 per million tokens, one flat rate across the whole window.

Google says Hydrogen scores 51 on the Intelligence Index from Artificial Analysis with thinking set to high, the default, one point above Argon at 50, and 54 at the top xhigh setting. It also says Hydrogen finishes tasks 2.5 times faster than Gemini 3.8 Flash at the default setting.