Google launches Gemini 4 Hydrogen, a faster, cheaper coding model

Google released Gemini 4 Hydrogen on November 12, 2026, a lower-cost, lower-latency model positioned as a daily-driver alternative to its frontier Gemini 4 Carbon model. The launch, detailed in a Google blog post, rolled out immediately across AI Studio, the Gemini API, Antigravity, Android Studio and other Google surfaces.

What's new in Gemini 4 Hydrogen

• Model ID gemini-4-hydrogen is a fast, efficient daily-driver model, distinct from the frontier Gemini 4 Carbon and Gemini 4 Argon models.
• Pricing is $1.50 per 1 million input tokens and $7.50 per 1 million output tokens, about 62 percent cheaper than Gemini 4 Carbon.
• The context window covers 1 million input tokens and up to 1 million output tokens.
• A new configurable thinking level called xhigh joins the existing low, medium and high levels.
• It scores 51 on the Artificial Analysis Intelligence Index at high and 54 at xhigh, ahead of Gemini 4 Argon's 50, and 57 percent on Terminal-Bench 4.1 versus 18.2 percent for Gemini 3.8 Flash.
• At the high setting it runs about 2.5 times faster than the prior generation and cuts token usage 12 to 23 percent on enterprise workflows.

What it means for builders

Hydrogen gives developers on the Gemini platform a cheaper, faster option for everyday coding and agentic work. Google also reports a more than fivefold reduction in fabricated claims and up to an 81 percent reduction in agentic overreach, with a 96.7 percent honesty rate on its internal evaluations, pointing to steadier behavior for agents that take actions rather than just generate text.

Availability

• Rollout began globally on November 12, 2026 across Google AI Studio, the Gemini API, Android Studio, the Gemini app, Google Search, Gemini for Business and Google Antigravity, per the AI Studio changelog.
• In Antigravity, a new Orchestrator feature dynamically switches between Carbon and Hydrogen within a single session to cut token cost while aiming to preserve Carbon-level quality.
• Vercel added the model to its AI Gateway the same day under the ID google/gemini-4-hydrogen, succeeding Gemini 3.8 Flash at similar latency with no markup pricing, according to Vercel's changelog.

Sources

• Introducing Gemini 4 Hydrogen (Google, November 12, 2026, official)
• AI Studio changelog, Nov 12 2026 (Google AI Studio, November 12, 2026, official)
• Gemini 4 Hydrogen now available on AI Gateway (Vercel, November 12, 2026, official)

Written with AI assistance, strictly from the sources listed above. Before it went live, automated checks verified its dates, sources and every language edition.