Tutorials & Tips

Gemini 4 Hydrogen: $1.50 per 1M input

Google DeepMind released Gemini 4 Hydrogen on November 12, 2026. On the official pricing page, the standard paid rate is $1.50 per million input tokens and $7.50 per million output tokens, including thinking tokens, at a flat rate across the 1 million-token context window. Unlike Carbon and Argon, Hydrogen also has a free tier.

The API name is gemini-4-hydrogen. Officially, it is the fast execution model for everyday coding and agentic workflows. Gemini 4 Carbon and Gemini 4 Argon remain the models for deep, long-horizon reasoning.

How much cheaper is it than Carbon and Argon?

These are the official standard prices still in effect today, in U.S. dollars per million tokens:

  • Hydrogen: $1.50 input, $7.50 output (including thinking tokens). The pricing page marks free-tier input and output as free of charge. The page does not list a price-increase date.
  • Carbon and Argon: $2 input, $10 output (including thinking tokens). Neither is on the free tier. From December 1, 2026, they rise to $4 input and $20 output. The official page does not say which time zone that change uses, and this article will not invent a clock time.

So Hydrogen’s standard price is already lower than Carbon and Argon. After December 1, if Hydrogen’s listed price is unchanged, the gap will be wider. That is a comparison with today’s page, not a promise the price will never change.

Hydrogen’s Batch and Flex paid rates are $0.75 input and $3.75 output; those tiers are not on the free tier. Priority is higher: $2.70 input and $13.50 output. On all three Gemini 4 models, the output price includes thinking tokens, so the bill is not just the visible reply. Free-tier content may be used to improve Google’s products; the paid standard tier says it will not. Check rate limits on the official page when you use it.

How should you choose it day to day?

Our suggestion is simple: for everyday coding, bug fixes, and agent workflows, start with Hydrogen, especially if you are on the free tier or watching the bill. For hard, long-horizon problems, use Carbon or Argon. We already compared those two, and the December 1 increase, here: How to choose Gemini 4 Carbon and Argon.

Hydrogen has four thinking levels: low, medium, high (the default), and a new xhigh tier. Google also introduced Orchestrator in Google Antigravity, which can switch between Carbon and Hydrogen in the same session. DeepMind says that on Terminal-Bench 4.0 it reached near-Carbon task success rates while, on average, using 24% less quota and finishing 19% faster. Those are official results, not our own tests.

Officially, Hydrogen is rolling out to the Gemini API, Google AI Studio, the Gemini app, Google Search, Gemini for business, and paid Google AI Pro and Ultra users in Antigravity. In the Gemini app, free users may get Hydrogen through automatic model selection when the system judges it appropriate. Whether those surfaces open in your country, and whether you must sign in or add a card, should be checked on the official pages themselves.

Our take

DeepMind also says Hydrogen scores 51 on the Artificial Analysis Intelligence Index at its default high thinking level, just above Argon’s 50, and 54 at xhigh. That is a one-point gap at the default setting. We would not tell you, on that basis, that Hydrogen is already smarter than Argon and can replace it on the hardest problems. The more reliable guide is Google’s own split: use Hydrogen for everyday execution, and keep Carbon or Argon for the deepest long-horizon work.

Sources:

Quick Facts

Published
Source
Google DeepMind / Google AI Studio
Category
Tutorials & Tips
Topic
Model ReleasesTutorials & Practice
Why it matters

Gemini 4 Hydrogen 官方标准价每百万输入 1.50 美元、输出 7.50 美元,有免费层;日常写代码先用它。

Topics

Related entities

How to Cite

When citing this article, keep the title, date, and permalink, and credit Orange News as the source.

Orange News: Gemini 4 Hydrogen: $1.50 per 1M input. November 13, 2026. https://new.nat6.net/en/news/custom-6fb5a47b9786421d/

View all →