Google ships Gemini 4 Hydrogen at $1.50 per million input tokens

Google DeepMind introduced Gemini 4 Hydrogen on Nov 12, 2026 and describes it as the model for everyday agentic and coding work. It sits below the company's frontier models, Gemini 4 Carbon and Gemini 4 Argon, which Google positions for long, difficult reasoning problems.

Price and access

Hydrogen costs $1.50 per million input tokens and $7.50 per million output tokens. The price is the same across the whole 1 million-token context window. It accepts up to 1 million input tokens and can return up to 1 million output tokens, and it handles PDFs, spreadsheets, charts, audio and recorded video.

It is available in all tiers of Google AI Studio, in the Gemini API and in Gemini for business, through the API or the model drop-down. It also appears in the Gemini app and in Google Search. In the Gemini app, free users reach it through automatic model selection when it suits the prompt. Gemini in Google Sheets gets it for eligible business subscriptions, and for Google AI Pro and Ultra subscribers.

Four thinking levels

The model has four settings: low, medium, high (the default) and a new xhigh tier. Google says quality, latency and cost scale predictably between them, so a task only pays for the reasoning depth it needs.

What Google reports

All figures below are Google's own.

• On the Artificial Analysis Intelligence Index, Hydrogen scores 51 at the default level against 50 for Argon, and 54 at xhigh.
• Against Gemini 3.8 Flash, it finishes tasks 2.5 times faster at the default level and scores 57% on Terminal-Bench 4.1, up from 18.2%. Google also says fabricated claims fell by more than 5x.
• Partners testing SQL and data-heavy enterprise workflows saw runs 2x to 2.8x faster, with 12% to 23% fewer tokens than 3.8 Flash.
• It ranks first on the Vals Tax Agent Bench (52.3%), ties for first on the Vals Legal Research Bench (55.8%) and leads its class on AutomationBench-AA.

Multi-model setups

Google Antigravity gets a new Orchestrator that switches between models such as Carbon and Hydrogen inside one session. On Terminal-Bench 4.0, Google says it came close to Carbon's success rate while using 24% less quota and finishing 19% faster. For parallel agents, Carbon can coordinate while Hydrogen runs the worker and verifier agents. Antigravity access to Hydrogen is for paid Google AI Pro and Ultra subscribers.

Safety claims

Google says Hydrogen is more than 2.3x more resilient to indirect prompt injection on the Gray Swan IPI benchmark. It also reports that overreach in agent tasks falls by up to 81% against 3.8 Flash, and that the model is up to 2x less likely to bypass permission boundaries when a task is blocked. These are vendor evaluations, not independent results.