ChatModelsComparePricingDocs
CouncilClaude CodeCursorCheapest modelsBest for codingFor agentsAgent templatesFree models

LiveAdded 12 Nov 2026

Gemini 4 Hydrogen API

Google DeepMind's fast, efficient Gemini 4 model for everyday coding and agentic work: four thinking levels and a 1M-token window at one flat price.

Open playgroundView API docs

Model spec

Input
$1.50/ 1M tokens
Output
$7.50/ 1M tokens
Cached input
$0.075/ 1M tokens
Context
1.05Mtokens
Max output
66Ktokens

Model ID

OpenAI-compatible · api.companyfabric.com/v1

Overview

About Gemini 4 Hydrogen

Gemini 4 Hydrogen is Google DeepMind's fast, efficient Gemini 4 model, built for everyday coding and agentic work: rapid execution across tool-using agents, iterative repository debugging, data analysis and interactive app building. Google announced it and released it in the Gemini API on 12 November 2026, as the execution model beside Gemini 4 Carbon and Gemini 4 Argon, which go deeper on the hardest problems. It reads text, images, audio, video and PDFs, works in a 1,048,576-token window shared by input, thinking and output, and has four thinking levels in Google's API: low, medium, high (the default) and xhigh. It is priced flat across the whole window at $1.50 input and $7.50 output per million tokens, with cached input at $0.075, and there is no introductory price to expire. On CompanyFabric it runs on Google's paid Gemini API, on the same key and balance as every other model, with no Google account to set up. Every model on CompanyFabric shares the same API key, billing balance, and rate-limit envelope, one integration covers the entire library.

Facts

At a glance

Availability

Live

Added 12 Nov 2026

Pricing

$1.50 in · $7.50 out

Per 1M tokens, metered exactly per call

Context window

1.05M tokens

Up to 66K tokens out

Commercial use

See license

Commercial use per Gemini API paid-tier terms; Google does not use paid-tier prompts to train its models.

Variants

All endpoints

Gemini 4 Carbon

$2/Mtok in

Google DeepMind's strongest reasoning model, built for sustained agentic coding, multimodal knowledge work, scientific reasoning and cybersecurity defense. Introductory price until 30 November.

Chat Completions

Gemini 4 Argon

$2/Mtok in

Google DeepMind's frontier model for long-horizon coding, legal, finance and science work, #1 on the Vals Index at launch. Introductory price until 30 November.

Chat Completions

Gemini 3.7 Flash

$0.75/Mtok in

Google's current fast model, million-token context.

Chat Completions

Gemini 3.5 Flash

Retired$1.5/Mtok in

The previous stable Flash, still current for most work.

Chat Completions

Gemini 2.5 Pro

Retired$1.25/Mtok in

Google's stable Pro tier, the non-preview reasoning model.

Chat Completions

Gemini 2.5 Flash

Retired$0.3/Mtok in

Long-context workhorse on the stable 2.5 line.

Chat Completions

Gemini 3.1 Pro (preview)

Preview$2/Mtok in

Google's newest Pro tier, shipped as a preview.

Chat Completions

Gemini 3 Flash (preview)

Preview$0.5/Mtok in

The Gemini 3 Flash preview line.

Chat Completions

Gemini 3.8 Flash

$0.75/Mtok in

Google's most capable Flash model, built for long-horizon software engineering and autonomous agents.

Chat Completions

Gemini 3.6 Flash

$0.75/Mtok in

The previous-generation Flash, balancing speed and multimodal work across everyday agentic tasks.

Chat Completions

Gemini 3.5 Flash-Lite

$0.3/Mtok in

Cost-efficient Gemini for high-volume agentic work, translation and simple data processing.

Chat Completions

Gemini 3.1 Flash-Lite

$0.25/Mtok in

Google's cheapest text model here, for high-volume classification and extraction.

Chat Completions

Gemini 4 Hydrogen

$1.5/Mtok in

Google DeepMind's fast, efficient Gemini 4 model for everyday coding and agentic work: four thinking levels and a 1M-token window at one flat price.

Chat Completions

Quickstart

How to use the Gemini 4 Hydrogen API

  1. 1

    Create a free CompanyFabric account, starter credits are included, no card required.

  2. 2

    Copy your API key from the dashboard. One key covers every model in the library.

  3. 3

    Point any OpenAI SDK at api.companyfabric.com/v1 and set model to "google/gemini-4-hydrogen".

  4. 4

    Read the exact metered cost of the call from the response, the same number shown in the playground.

Use cases

What teams build with it

Coding agents and test loops

Writing code, running tests and verifying changes in a tool-using loop, where speed per turn matters more than the deepest reasoning. Google says Hydrogen batches file reads and searches in parallel and makes focused edits instead of rewriting unrelated files.

coding agenttool useiteration

Worker agents beside a stronger planner

Plan on Carbon or Argon and execute on Hydrogen, switched by changing the model id on the same key and balance, so the dearer model only runs where judgment is needed.

multi-modelagentscost

Data and document workflows

Reading dense PDFs, spreadsheets and charts into structured, cited reports, and high-volume analysis where a million-token window at one flat price keeps cost predictable.

multimodaldata analysisdocuments

Get better output

Prompting tips

  • Leave room for thinking: max_tokens caps thinking and answer together, and thinking tokens bill as output, so set it high enough that the final answer is not cut off.
  • Send the whole task. Google reports Hydrogen flags missing context instead of guessing (more than five times fewer fabricated claims than Gemini 3.8 Flash), so give it the files and constraints and let it say what it still needs.
  • Do not bother sending temperature, top_p or top_k: Google's server ignores custom values for Gemini 4 Hydrogen.

Exact, metered, prepaid

Gemini 4 Hydrogen pricing

Gemini 4 Hydrogen pricing is per token: $1.5 / M input · $7.5 / M output · $0.075 / M cached input. Output tokens cost 5× input tokens, so long generations, not long prompts, are what actually drive most bills. A typical call with a 2k-token prompt and a 1k-token answer costs about $0.0105.

Every call is metered exactly and attributed to your key, the dashboard, the API usage response, and your invoice all read from the same ledger row. Cap any key's monthly spend with a budget cap, and bring your own provider key (BYOK) at 0% platform fee if you already have a direct contract.

EndpointTypePrice
Gemini 4 CarbonChat$2 / M input · $10 / M output · $0.1 / M cached input
Gemini 4 ArgonChat$2 / M input · $10 / M output · $0.1 / M cached input
Gemini 3.7 FlashChat$0.75 / M input · $3.75 / M output
Gemini 3.5 FlashChat$1.5 / M input · $9 / M output
Gemini 2.5 ProChat$1.25 / M input · $10 / M output
Gemini 2.5 FlashChat$0.3 / M input · $2.5 / M output
Gemini 3.1 Pro (preview)Chat$2 / M input · $12 / M output
Gemini 3 Flash (preview)Chat$0.5 / M input · $3 / M output
Gemini 3.8 FlashChat$0.75 / M input · $3.75 / M output
Gemini 3.6 FlashChat$0.75 / M input · $3.75 / M output
Gemini 3.5 Flash-LiteChat$0.3 / M input · $2.5 / M output
Gemini 3.1 Flash-LiteChat$0.25 / M input · $1.5 / M output
Gemini 4 HydrogenChat$1.5 / M input · $7.5 / M output · $0.075 / M cached input

Starter credits

Use Gemini 4 Hydrogen free

Your CompanyFabric starter credits cover your first Gemini 4 Hydrogen calls at no cost, create a free account, no card required, and the credits apply to Gemini 4 Hydrogen and every other model on the key. When they run out, prepaid credits or BYOK (0% fee) take over with no plan to pick.

All ways to use AI models free →

Production-grade

Code samples

import OpenAI from "openai"

const client = new OpenAI({
  baseURL: "https://api.companyfabric.com/v1",
  apiKey: process.env.COMPANYFABRIC_API_KEY,
})

const res = await client.chat.completions.create({
  model: "google/gemini-4-hydrogen",
  max_tokens: 1024,
  temperature: 0.7,
  messages: [{ role: "user", content: "Say hello from the fabric." }],
})

console.log(res.choices[0].message.content)

Playground

Try it in the browser

Playground$1.5 in / $7.5 out per 1M tokens

Comparisons

Gemini 4 Hydrogen vs the alternatives

Gemini 4 Hydrogen vs Claude Sonnet 5.5

Claude Sonnet 5.5 is $2 / $10 per million tokens and Hydrogen is $1.50 / $7.50, a quarter cheaper per token, with a 1,048,576-token window against Sonnet's 1,000,000 and the price flat across all of it. Google reports Hydrogen matching or beating models in its class on coding and agent benchmarks; those are Google's numbers, not independent results. Run your own agent loop on both through one key and compare the answers and usage.cost_usd.

Full comparison: Gemini 4 Hydrogen vs Claude Sonnet 5.5 →

Gemini 4 Hydrogen vs GPT-6 Sol

GPT-6 Sol is $2 / $10 per million tokens against Hydrogen's $1.50 / $7.50, and both work in a window of about a million tokens. Hydrogen is a quarter cheaper per token; whether it is cheaper per task depends on how many thinking tokens each spends. The same key reaches both, so measure on your own workload.

Full comparison: Gemini 4 Hydrogen vs GPT-6 Sol →

Gemini 4 Hydrogen vs Gemini 3.8 Flash

Gemini 3.8 Flash is $0.75 / $3.75 per million tokens through the end of 2026 and $1.50 / $7.50 after, which is Hydrogen's price, so Hydrogen costs twice as much per token until then and the same after. Google positions it as the upgrade: 2.5 times faster at its default setting, Terminal-Bench 4.1 at 57% against 18.2%, and more than five times fewer fabricated claims. Those are Google's numbers. If your workload is simple and cost-sensitive, keep 3.8 Flash while its introductory price lasts; for agent and coding loops, test Hydrogen.

Full comparison: Gemini 4 Hydrogen vs Gemini 3.8 Flash →

Compare

Gemini 4 Hydrogen head to head

Gemini 4 HydrogenvsClaude Sonnet 5.5Gemini 4 HydrogenvsGPT-6 SolGemini 4 HydrogenvsGemini 3.8 Flash

FAQ

Frequently asked questions

How do I get a Gemini 4 Hydrogen API key?

Create a CompanyFabric account, and one key unlocks Gemini 4 Hydrogen and every other model in the library. Free starter credits are included, no subscription, no card required to try it.

Why use Gemini 4 Hydrogen on CompanyFabric instead of going direct?

One key, one prepaid balance, and one rate-limit envelope across the whole library, at pricing at parity with or below the direct API. Model switching is a one-line change, and BYOK routing is 0% fee.

Is the API OpenAI-compatible?

Yes, point any OpenAI SDK at api.companyfabric.com/v1 and set model to "google/gemini-4-hydrogen".

Can I use the output commercially?

Commercial use per Gemini API paid-tier terms; Google does not use paid-tier prompts to train its models.

How does billing work?

Prepaid credits via card, or BYOK with your own provider keys at 0% platform fee. No subscription required; balances never expire.

How much does the Gemini 4 Hydrogen API cost?

$1.50 per million input tokens, $7.50 per million output tokens and $0.075 per million cached input tokens, the same at every prompt length across the whole 1,048,576-token window. Thinking tokens bill as output. There is no introductory price to expire. Each response carries what it cost in usage.cost_usd.

How does Gemini 4 Hydrogen compare with Gemini 4 Carbon and Argon?

Google positions Carbon and Argon for deep, long-horizon reasoning on your hardest problems and Hydrogen for rapid execution across everyday coding and agent workflows. Google reports Hydrogen at 51 on the Artificial Analysis Intelligence Index at its default high thinking level, against 50 for Argon, and 54 at its highest xhigh setting. Per token it is cheaper than both: $1.50 / $7.50 against $2 / $10 for Carbon and Argon, which rise to $4 / $20 on 1 December. These are Google's numbers, not independent results.

What are Gemini 4 Hydrogen's thinking levels?

Google's API has four: low, medium, high (the default) and a new xhigh, and it rejects minimal with an error. Higher levels spend more thinking tokens, which bill as output, so they cost more and take longer. If a request sets a level Google does not accept, the 400 comes back with Google's reason.

How long can a Gemini 4 Hydrogen answer be?

Up to 65,536 output tokens per request, inside a 1,048,576-token window that input, thinking and output share. Google's long decode can go further, up to the whole window, but it needs Google's own continuation protocol, which CompanyFabric does not expose yet.

Do I need a Google account to use Gemini 4 Hydrogen?

No. Google also offers Hydrogen on its own free tier and in the Gemini app, but through CompanyFabric you call gemini-4-hydrogen with your CompanyFabric key on Google's paid API tier, where Google does not train on your prompts, billed to your prepaid balance.

Provider

About Google

Google publishes the gemini family. Also in this family: Gemini 4 Carbon, Gemini 4 Argon, Gemini 3.7 Flash, Gemini 3.5 Flash, Gemini 2.5 Pro, Gemini 2.5 Flash, Gemini 3.1 Pro (preview), Gemini 3 Flash (preview), Gemini 3.8 Flash, Gemini 3.6 Flash, Gemini 3.5 Flash-Lite, Gemini 3.1 Flash-Lite. License: Commercial use per Gemini API paid-tier terms; Google does not use paid-tier prompts to train its models.

One key. Every model. Exact prices.

Run Gemini 4 Hydrogen with free starter credits. Free starter credits included, no subscription required.

Get your API keyBrowse all models