Google launched Gemini 4 Hydrogen on 12 November, the fast, lower-cost model in its Gemini 4 family, at $1.50 per million input tokens and $7.50 per million output tokens, according to Google’s announcement and its Gemini API pricing page. That is twice the $0.75 and $3.75 Google charges for Gemini 3.8 Flash through 31 December, by our arithmetic.
The official rates match, to the cent, the Google Cloud catalogue rows we reported at 08:55 UTC, about nine hours before Google DeepMind chief executive Demis Hassabis posted on X at 18:11 UTC that Google was “Very excited to release Gemini 4 Hydrogen”.
Business Pill · Usage-Based Pricing Explained
An evergreen Business Engineer explainer on usage-based pricing, paying for work done rather than for seats, the model behind per-token API rates. The video teaches the general idea only and does not discuss Gemini or Hydrogen.
The key insight: As we read the pricing page, Google has priced Hydrogen at the rate Gemini 3.8 Flash moves to on 1 January 2027, so the step up from Flash is a model change today and a price change later: $1.50 and $7.50 per million tokens either way.
Google says Hydrogen is “built for rapid execution across diverse agentic workflows”, while its frontier models, Gemini 4 Carbon and Gemini 4 Argon, “excel at deep, long-horizon reasoning on your hardest problems”. The model id is gemini-4-hydrogen, and its model page lists a context window of 1,048,576 tokens, with output of 65,536 tokens by default and up to 1,048,576 with long decode.
The blog says Hydrogen is “priced flat across its entire 1 million-token context window”. It is rolling out to all tiers of Google AI Studio, the Gemini API, Gemini for business, the Gemini app and Google Search, and to paid Google AI Pro and Ultra subscribers in Google Antigravity, Google says. The pricing page lists a free tier for Hydrogen; Carbon and Argon are “not available on the Free tier”.
Cost of one job of 1 million input and 200,000 output tokens at each model’s standard API rates on Google’s, OpenAI’s and Anthropic’s pricing pages, read 12 November 2026. Totals are our arithmetic.
Demis Hassabis, chief executive of Google DeepMind, posted this at 18:11 UTC on 12 November 2026, linking to Google’s launch blog. Gennaro passed it to the desk. The piece quotes its opening line; the prices and benchmarks come from Google’s blog and pricing page.
Introducing Gemini 4 Hydrogen, an extremely capable and fast model built to bring Gemini 4 to everyone!
Available to all tiers of Google AI Studio, Gemini API, Gemini for business, the Gemini app, Google Search, and paid Google AI Pro + Ultra subscribers in Google Antigravity.
Logan Kilpatrick of Google posted this at 18:05 UTC on 12 November 2026, the post Hassabis quoted. It lists where Hydrogen is available, the same surfaces the piece takes from Google’s blog.
The Prices, by Our Arithmetic
At $1.50 and $7.50, Hydrogen costs 75% of Carbon and Argon at their current $2 and $10, and 37.5% of the $4 and $20 the pricing page lists for both from 1 December. Hydrogen’s rates equal the $1.50 and $7.50 the page lists for Gemini 3.8 Flash from 1 January 2027. Batch and Flex rates for Hydrogen are $0.75 and $3.75, half the standard rate.
Against rivals’ standard rates, Hydrogen is 1.5 times OpenAI’s GPT-6.1 Terra at $1 and $5, per OpenAI’s pricing page. It is 75% of GPT-6.1 Sol and Anthropic’s Claude Sonnet 5.5 at $2 and $10, and 37.5% of Claude Opus 5.5 at $4 and $20, per Anthropic’s pricing page. On one job of 1 million input and 200,000 output tokens, that is $3.00 for Hydrogen against $2.00, $4.00 and $8.00, by our arithmetic.
The Benchmarks Google States
Google says Hydrogen scores 51 on the Artificial Analysis Intelligence Index at its default high thinking level, against 50 for Argon, and reaches 54 at its highest xhigh setting. Against Gemini 3.8 Flash, the blog says Hydrogen scores 57% on Terminal-Bench 4.1 versus 18.2%, finishes tasks 2.5 times faster at its default setting and cuts fabricated claims by more than 5x. The Terminal-Bench gap is 38.8 points, or 3.1 times, by our arithmetic.
Google also lists 58.8% on HLE-Diamond, 52.3% on Vals Tax Agent Bench, 55.8% on Vals Legal Research Bench, 53.4% on the new Multimodal Knowledge Agent benchmark and 71.6% on Agents’ Last Exam v1.1. It introduced Orchestrator in Google Antigravity, which switches between Carbon and Hydrogen in one session; on Terminal-Bench 4.0, Google says, it reached “near-Carbon task success rates” while using 24% less quota and finishing 19% faster on average.
As we read the launch post, Google is selling Hydrogen less as a cheaper Carbon than as the worker in a two-model setup, with Orchestrator switching between them; the post states 24% less quota for that setup on Terminal-Bench 4.0.
As we read the two price lists together, Hydrogen sits between OpenAI’s GPT-6.1 Terra and GPT-6.1 Sol on standard rates, at 1.5 times the first and 75% of the second, by our arithmetic.
Google, Introducing Gemini 4 Hydrogen, 12 November 2026
“Hydrogen is built to stand on its own as a model for everyday work — or to pair with Carbon and Argon as the high-efficiency execution engine in multi-model workflows.”
Three Implications
FOR DEVELOPERS ON GEMINI 3.8 FLASH Hydrogen costs twice Flash’s rate through 31 December and the same rate from 1 January 2027, per Google’s pricing page, by our arithmetic.
FOR TEAMS ON CARBON OR ARGON Hydrogen costs 75% of their rates now and 37.5% from 1 December, by our arithmetic; Google says Hydrogen scores 51 against Argon’s 50 on the Artificial Analysis index.
FOR BUYERS COMPARING LABS At standard rates Hydrogen is 1.5 times GPT-6.1 Terra and 37.5% of Claude Opus 5.5, per OpenAI’s and Anthropic’s pricing pages, by our arithmetic.
The framework describes the token as “the unit of cognition the model produces, the unit of compute the data center serves, the unit of price the lab charges, and the unit of value the enterprise extracts.”
As we read it, Hydrogen’s launch is a move on the price unit: the same Gemini 4 family, sold at 37.5% of Carbon’s 1 December rate per token; the framework predates the launch and does not discuss Hydrogen.
Terminal-Bench 4.1 scores as stated in Google’s launch post: 57% for Gemini 4 Hydrogen against 18.2% for Gemini 3.8 Flash. The figures are Google’s own; the 3.1x ratio is our arithmetic.
What Is Not Established
The launch blog, which we read in full, gives no head-to-head score for Hydrogen against Carbon on the same benchmark, and no Artificial Analysis score for Carbon. The benchmark figures are Google’s own and were not independently checked for this piece. The blog does not say how much traffic in the Gemini app or Search Hydrogen will carry, or what it costs Google to serve. Google was not contacted for this piece.
This Week In The AI Supercycle
The Great Convergence of Capital, Power, and Geopolitics
This week’s issue: for two years the AI build-out was priced as a supply story, and last week as a credit story. This week it was priced as a permission story, and the answer arrived as a cost, not a ban.
Gemini 4 Hydrogen launched at $1.50 and $7.50 per million tokens, the same rates we found in Google’s Cloud catalogue nine hours earlier. That is twice Gemini 3.8 Flash’s price through 31 December, 75% of Carbon today and 37.5% from 1 December, and 1.5 times GPT-6.1 Terra, by our arithmetic; Google says it scores 51 against Argon’s 50 on the Artificial Analysis index.
95,000+ executives read Business Engineer for the AI strategy frameworks cited by ChatGPT, Claude, and Perplexity.
A note on sourcing. We read Google’s launch post of 12 November 2026 in full, and took prices, the free tier, the model id and token limits from the Gemini API pricing and model pages, and rival rates from OpenAI’s and Anthropic’s pricing pages, all read on 12 November. Benchmarks are Google’s own. Ratios and workload totals are our arithmetic. Google was not contacted for this piece.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
