Gemini 4 Hydrogen is Google’s newest AI model, and it is rolling out now to the Gemini app, Google Search and developers. Announced on November 12, it is built for speed and everyday agent work rather than the deepest reasoning, yet Google says it edges past one of its own frontier models on a key intelligence index.
Hydrogen joins the Gemini 4 family as the fast, low-cost option. Google says it often finishes tasks about 2.5 times faster than Gemini 3.8 Flash at its default setting, at a price of $1.50 per million input tokens.
In this article
What is Gemini 4 Hydrogen?
Google now splits Gemini 4 into roles. Carbon and Argon are its frontier models for long, hard reasoning problems. Hydrogen is tuned for quick execution in coding and agentic workflows, and it can run on its own or alongside the bigger models.
In Google Antigravity, a new Orchestrator feature can switch between Carbon and Hydrogen within one session. Carbon can coordinate a job while Hydrogen runs the worker and verifier agents, according to Google’s announcement.
Gemini 4 Hydrogen benchmarks and price
These are Google’s own figures:
| Measure | Gemini 4 Hydrogen |
|---|---|
| Artificial Analysis Intelligence Index | 51 at default high thinking (Argon: 50), 54 at xhigh |
| Terminal-Bench 4.1 | 57% (Gemini 3.8 Flash: 18.2%) |
| HLE-Diamond | 58.8% |
| Vals Tax Agent Bench | 52.3%, ranked first overall |
| Price | $1.50 input, $7.50 output per million tokens |
| Context | Up to 1 million tokens in and out |
The model has four thinking levels: low, medium, high (the default) and a new xhigh tier. Pricing stays flat across the full 1-million-token window, which matters for long documents and large codebases.
Google also claims Hydrogen cuts fabricated claims by more than five times compared with Gemini 3.8 Flash. In partner tests on SQL, graph and data-heavy enterprise jobs, it ran 2 to 2.8 times faster and used 12% to 23% fewer tokens.
Where you can use Gemini 4 Hydrogen
The rollout covers consumers, subscribers and developers:
Gemini app: Hydrogen appears in the model dropdown. Free users get it through automatic model selection when it suits the prompt.
Google Search: It is coming to Search for all users worldwide. Google calls it “the biggest model upgrade to Search yet.”
Google Sheets: Available to Google AI Pro and Ultra subscribers, where Google says outputs arrive more than 30% faster.
Developers: All tiers of the Gemini API in Google AI Studio, plus Android Studio. AI Pro and Ultra subscribers can also use it in Antigravity.
In Search, Google says Hydrogen handles longer, multi-part questions and typos better, and runs parallel queries across subtopics to surface a wider range of websites. Answers can include more images, comparison tables and custom tools, according to the Google Search blog.
Why Gemini 4 Hydrogen matters
Most people never pick a model. They type into Search or the Gemini app and get whatever Google serves. Putting Hydrogen behind those products means the upgrade reaches a huge audience without anyone changing a setting.
For developers, the pitch is price and speed. At $1.50 per million input tokens, Hydrogen undercuts the frontier tier, and Google says it scores just above Argon on the Artificial Analysis index. That makes it a candidate for agent loops that call a model hundreds of times.
Google also published safety claims, including up to 81% less overreach in agent tasks and 2.3 times more resilience to indirect prompt injection on the Gray Swan benchmark. Like the benchmark scores, these come from the company, and independent testing will show how they hold up. The AI race keeps moving fast; Anthropic shipped its own upgrade days ago, as we covered in our look at Claude Fable 5.2. Google also brought Gemini Live to CarPlay this week, and you can find more in our latest stories.
FAQ
Is Gemini 4 Hydrogen free?
Free Gemini app users can get it through automatic model selection, and it powers Google Search for everyone. Some uses, such as Sheets and Antigravity, need a Google AI Pro or Ultra plan.
How much does Gemini 4 Hydrogen cost for developers?
$1.50 per million input tokens and $7.50 per million output tokens, flat across its 1-million-token context window.
What is the difference between Gemini 4 Hydrogen and Gemini 4 Carbon?
Carbon is built for deep, long-horizon reasoning on hard problems. Hydrogen is faster and cheaper, aimed at everyday coding and agent tasks.
Sources: Google, Google Search blog, Digital Trends
Leave a Reply