Justin McKelvey

Justin McKelvey

Ran snowboard shops. Taught myself to code. Now I build AI that does real work.

• AI for Business • 9 min read •

Gemini 4 Hydrogen: Who Gets It, What It Costs, and What Changes for Your Business (2026)

Quick Answer

Gemini 4 Hydrogen is Google's new fast everyday model, released November 12, 2026, and it's now behind Google Search for everyone. In the Gemini app, free users only get it when Gemini's auto selection decides a prompt needs it; Google AI Plus ($4.99 a month) is the cheapest plan that lists it by name. Pro ($19.99) also gets it in Sheets. On the API it's $1.50 per million input tokens and $7.50 output.

Read on blog.google and gemini.google, November 13, 2026 · Author: Justin McKelvey, AI consultant, former CTO at an AI startup, 50+ products shipped

TL;DR: "for everyone" has fine print

Google's headline says Hydrogen brings Gemini 4 "to everyone." Its own plans page still lists the Flash-Lite model on the free tier. Both are true, and the gap is the thing worth knowing before you tell your team "we have Gemini 4 now."

Here's the short version. Search got Hydrogen for every user, which matters more to your business than the chatbot does, because it changes how Google answers questions about your industry. The Gemini app gets it fully on paid plans and sometimes on free. Developers pay a flat, low rate. Below: who gets what, how to use it on a real owner task, what it costs to automate with, and what to watch.

What Gemini 4 Hydrogen is, and who gets it

Hydrogen is the fast model in Google's Gemini 4 family. The two bigger models, Carbon and Argon, are for long, hard reasoning. Hydrogen is for everyday work: writing, summarizing, comparing documents, working across Gmail and Drive, and running the routine steps in automations. It replaces Gemini 3.8 Flash and is the first of these models without "Flash" in the name.

Google's claims, all from its own announcement: 51 on the Artificial Analysis Intelligence Index at the default setting (Argon scores 50), 54 at the new xhigh setting, tasks finished 2.5 times faster than 3.8 Flash, and fabricated claims cut by more than five times. It reads up to 1 million tokens at once (several hundred pages) and can write up to 1 million back.

Where Who gets Hydrogen Price (Google pages, Nov 13, 2026)
Google Search (AI answers) All users, globally $0
Gemini app, free Only via auto model selection; the plan lists Flash-Lite $0
Gemini app, Google AI Plus By name, 2x the free usage $4.99/month
Gemini app, Google AI Pro By name, plus Argon; Hydrogen in Sheets and Antigravity $19.99/month
Gemini app, Google AI Ultra Highest access, plus Carbon $99.99 or $199.99/month
Google Workspace / Gemini for business Model menu, API, and Sheets on eligible plans Per seat
Gemini API and AI Studio All tiers, free tier included $1.50 in / $7.50 out per 1M tokens

How much you can actually use on each plan is a separate question; the caps are on Gemini usage limits.

How to use Gemini 4 Hydrogen, step by step

  1. Open the Gemini app and check the model menu. On Plus, Pro or Ultra, choose Gemini 4 Hydrogen. On the free plan you can't choose it; Gemini picks it for you when it thinks the prompt needs it.
  2. Connect what it should read. Google says Hydrogen is better at pulling from your uploads, Gmail, Google Drive and Connected Apps like Airtable and Dropbox. Connect only the ones you want in the answer.
  3. Give it the rules up front. Google's pitch is that Hydrogen follows tone, format and rules more closely. Test that: "Short list, no jargon, every price from the attached sheet, nothing invented."
  4. Use it in Sheets if you're on Pro or a business plan. Google says Gemini in Sheets gets Hydrogen first for Pro, Ultra and eligible business subscriptions.
  5. Raise the thinking level only when it's wrong. Developers can set low, medium, high (default) or xhigh. Higher costs more time and tokens; most owner tasks don't need it.

A worked example: three supplier quotes in one table

The numbers below are illustrative, not from a real business.

The situation. A 12-person landscaping company is choosing a supplier for next season's stone and mulch. Three quotes arrive as PDFs with different units, delivery terms and fine print.

The prompt. "I uploaded three supplier quotes. Put them in one table: item, quantity, unit price, total, delivery date, payment terms. Convert everything to price per ton. Flag anything one quote includes that the others don't, and quote the line it came from."

What good looks like. One table you can read in a minute, with the hidden differences called out (one supplier charges a fuel surcharge, one includes delivery, one wants 50% up front). The "quote the line" instruction matters: it's how you check the model didn't invent a term, which is exactly the failure Google says Hydrogen cuts by five times. Check it anyway.

The time math. Doing this by hand is 45 to 60 minutes of retyping. With Hydrogen it's about 10 minutes including checking every number against the PDFs. If the owner does this twice a month, that's roughly 1.5 hours back, on a $4.99 or $19.99 plan.

If you'd rather automate it. Say you run 2,000 customer emails a month through Hydrogen to sort and draft replies, at about 1,500 tokens in and 400 out each. That's 3 million input tokens ($4.50) and 800,000 output tokens ($6.00): about $10.50 a month on the API. The full rate card, including Carbon and Argon, is on Gemini API pricing.

What changes in Google Search, and why it matters for your website

This is the part most owners will feel without noticing. Google says Hydrogen is now behind Search for everyone, with better handling of long, multi-part questions, more tables and visuals in the answer, and an upgraded "query fan-out": Search runs several related searches at once and links "a broader range of helpful links than before."

Translation: when a customer asks Google a full question about your service, more of the answer is written by AI, and the sites it links are picked from a wider pool. That's a threat if your site doesn't answer the question plainly and a chance if it does. In this site's own Search Console data, 77% of Google impressions already happen inside an AI answer (our measurement, September 2026). Being one of the links inside that answer is the job now, and it's what answer engine optimization is about.

My take

I haven't run my own work on Hydrogen yet, so this is how I'd decide, not a review.

I tell clients to own the data and rent the model. Your customer list, your prices and your notes live in systems you control; the model is a utility you can swap. Hydrogen is a good reason to keep it that way, because Google shipped three Gemini 4 models in under four weeks (Argon in October, Carbon on November 10, Hydrogen on November 12). If your workflow only works with one model, you'll rebuild it every time Google ships.

For most owners, the useful move this week isn't upgrading. It's searching Google for the questions your customers ask and seeing whether your site shows up in the new answers. If you're paying for Gemini already, switch the model menu to Hydrogen for daily work and save the big models for the hard problems. If you're on the free plan, you don't need to buy anything to try it; you just can't count on getting it every time. I'll add hands-on notes here once I've used it on real work.

What to watch out for

  • Free isn't guaranteed. Free app users get Hydrogen only when auto selection picks it. Don't build a team process on the free plan and assume it's Gemini 4 every time.
  • Personal plans aren't business plans. Google AI Plus, Pro and Ultra attach to a personal Google account. If your company is on Workspace, look at Gemini for business, not personal subscriptions for staff. Whether a paid plan pays for itself is on is Gemini worth it.
  • "Fewer fabrications" isn't zero. Ask it to quote its source line for every number, and check the numbers before they reach a customer.
  • Connected apps widen what it reads. Connecting Gmail and Drive makes answers better and puts more of your data in the prompt. Decide on purpose, and keep patient or card data out unless your setup covers it.
  • Automations need a model plan. If you're building workflows on the API, design them so the model can be swapped. That's the first thing we check in AI implementation work.

Frequently asked questions about Gemini 4 Hydrogen

What is Gemini 4 Hydrogen?

Gemini 4 Hydrogen is Google's fast everyday model, released November 12, 2026. It replaces Gemini 3.8 Flash as the default workhorse and sits below the frontier models Gemini 4 Carbon and Argon. Google says it scores 51 on the Artificial Analysis Intelligence Index at its default setting (Argon scores 50), finishes tasks 2.5 times faster than 3.8 Flash, and makes fabricated claims more than five times less often.

Is Gemini 4 Hydrogen free?

Partly. In Google Search it's on for all users. In the Gemini app, Google says free users get Hydrogen through auto model selection when it's the right model for the prompt, and the free plan on Google's subscriptions page lists the Flash-Lite model. To pick Hydrogen by name you need a paid plan; Google AI Plus at $4.99 a month is the cheapest one that lists it (gemini.google, read November 13, 2026). Developers get a free tier in Google AI Studio.

How much does Gemini 4 Hydrogen cost on the API?

$1.50 per million input tokens and $7.50 per million output tokens, flat across the full 1 million-token context window, per Google's November 12, 2026 announcement. As a worked example, 2,000 customer emails a month at about 1,500 tokens in and 400 out each comes to roughly $10.50 a month.

Which plans get Gemini 4 Hydrogen in Google Sheets?

Google says Gemini in Google Sheets gets Hydrogen starting with Google AI Pro ($19.99 a month) and Ultra subscribers, and for eligible business subscriptions through Gemini for business.

Is Gemini 4 Hydrogen better than Gemini 4 Argon or Carbon?

It's faster and cheaper, not smarter at the hardest problems. Google positions Carbon and Argon for deep, long reasoning and Hydrogen for quick everyday work and as the worker model in multi-model setups. On the Artificial Analysis index Hydrogen scores 51 at default and 54 at its xhigh setting versus Argon's 50, so for most owner tasks Hydrogen is the right default.

Sources

  • Google, "Introducing Gemini 4 Hydrogen" (blog.google), published November 12, 2026, read November 13, 2026: availability across AI Studio, Gemini API, Gemini for business, Gemini app, Search, Antigravity (paid AI Pro and Ultra) and Sheets (AI Pro, Ultra, eligible business plans); footnote on free users and auto model selection; $1.50 / $7.50 per 1M tokens; benchmark and speed claims; four thinking levels.
  • Google, "Google Search gets a major upgrade with Gemini 4 Hydrogen" (blog.google), published November 12, 2026, read November 13, 2026: available globally in Search for all users; improved query fan-out and broader links; more visual and table responses.
  • Google AI plans (gemini.google/subscriptions), read November 13, 2026: Free $0 with Flash-Lite; Google AI Plus $4.99 with Gemini 4 Hydrogen, 2x Free usage; Google AI Pro $19.99 with Argon and Hydrogen, 4x Free; Google AI Ultra $99.99 (5x Pro) and $199.99 (20x Pro) with Carbon.
  • justinmckelvey.com Google Search Console, 28 days to September 1, 2026: share of impressions inside Google's AI features (our own measurement).
Free Resource Justin McKelvey

Get the Free AI Content Toolkit

The exact system I use to turn one idea into a month of content: atomization framework, voice template, prompt library, weekly system.

Frequently Asked Questions

What is Gemini 4 Hydrogen?
Gemini 4 Hydrogen is Google's fast everyday model, released November 12, 2026. It replaces Gemini 3.8 Flash as the default workhorse and sits below the frontier models Gemini 4 Carbon and Argon. Google says it scores 51 on the Artificial Analysis Intelligence Index at its default setting (Argon scores 50), finishes tasks 2.5 times faster than 3.8 Flash, and makes fabricated claims more than five times less often.
Is Gemini 4 Hydrogen free?
Partly. In Google Search it's on for all users. In the Gemini app, Google says free users get Hydrogen through auto model selection when it's the right model for the prompt, and the free plan on Google's subscriptions page lists the Flash-Lite model. To pick Hydrogen by name you need a paid plan; Google AI Plus at $4.99 a month is the cheapest one that lists it (gemini.google, read November 13, 2026). Developers get a free tier in Google AI Studio.
How much does Gemini 4 Hydrogen cost on the API?
$1.50 per million input tokens and $7.50 per million output tokens, flat across the full 1 million-token context window, per Google's November 12, 2026 announcement. As a worked example, 2,000 customer emails a month at about 1,500 tokens in and 400 out each comes to roughly $10.50 a month.
Which plans get Gemini 4 Hydrogen in Google Sheets?
Google says Gemini in Google Sheets gets Hydrogen starting with Google AI Pro ($19.99 a month) and Ultra subscribers, and for eligible business subscriptions through Gemini for business.
Is Gemini 4 Hydrogen better than Gemini 4 Argon or Carbon?
It's faster and cheaper, not smarter at the hardest problems. Google positions Carbon and Argon for deep, long reasoning and Hydrogen for quick everyday work and as the worker model in multi-model setups. On the Artificial Analysis index Hydrogen scores 51 at default and 54 at its xhigh setting versus Argon's 50, so for most owner tasks Hydrogen is the right default.

More on AI for Business

Copilot Notebooks: What You Get Free With Microsoft 365, What Needs Copilot, and the Sharing Trap (2026)

Copilot Notebooks lets you drop a project's files into one place and ask Copilot questions that are answered only from those files. It works with the Copilot Chat that comes with Microsoft 365 (up to 50 references), and the paid Microsoft 365 Copilot adds 300 grounded files plus Teams, SharePoint and OneDrive sources. How to set one up, a worked example, and the sharing setting most owners miss.

7 min

ChatGPT Business vs Plus (2026): Which One a Small Business Should Actually Pay For

ChatGPT Plus vs Business, as of September 2026: Plus is $20 a month for one person and OpenAI may train on your chats unless you opt out; Business is $20 a seat billed annually or $25 monthly, two seats minimum, with no training by default, admin controls, shared projects for groups, and the Company Knowledge plugin. The side-by-side from OpenAI's pages, and three owner scenarios with a verdict each.

8 min

ChatGPT for Customer Service (2026): What Works, What Breaks, and What It Costs a Small Business

ChatGPT for customer service, as of September 2026: it drafts replies, answers policy questions from your own documents, and triages an inbox for $20 to $25 a seat a month on ChatGPT Business. It can't look up an order, issue a refund, or remember a customer unless you connect the system that holds them. The honest capability line, the plan you need, a setup in six steps, the failure modes, and the point where a real agent takes over.

7 min

ChatGPT Pulse (2026): What It Was, Why OpenAI Retired It, and How to Get the Morning Brief for Your Business

ChatGPT Pulse, the daily research cards OpenAI previewed for Pro users in September 2025, was sunset on June 17, 2026 with a 14-day wind-down. What it did, which plans ever had it, why scheduled tasks replaced it, and how an owner rebuilds the morning brief on a Plus or Business seat without handing ChatGPT the wrong things.

7 min
Justin McKelvey, Austin, TX

Written by

Justin McKelvey

I spent 12 years running snowboard and mountain-bike shops in Colorado, then taught myself to code. Since then: 50+ products shipped, Head of Product at a national sports platform, CTO at an AI startup. Now I put AI to work in real businesses in Austin, TX and remote, and write down what actually works.

Work with me

Want more help? Get the free toolkit · Book a strategy call