Skip to main content

google/gemini-4-hydrogen/chat

Google's Gemini 4 Hydrogen chat endpoint handles everyday coding and agent steps, taking text and images and offering four thinking levels within a 1,048,576-token shared context.

Input $1.3247Output $6.6233per 1 million tokensCached input $0.0662

Workbench

Gemini 4 Hydrogen

What can Gemini 4 Hydrogen help with?

Missing the model you need?

FAST EVERYDAY REASONING API

Gemini 4 Hydrogen API at a glance

Everyday coding and agent work from Google, with thinking you can dial from low to xhigh.

Shared context
1M
1,048,576 tokens for input, thinking and reply together
Thinking levels
4
low, medium, high and xhigh on one field
Reply ceiling
64K
Up to 65,536 tokens per call, thinking included

Results

What Gemini 4 Hydrogen is good at

  • Positioning

    Sized for the daily loop

    Google positions Hydrogen for repository edits, debugging and agent steps that run many times a day, where a quick answer matters more than the deepest possible search.

  • Thinking

    A dial with four stops

    Set reasoning_effort to low for a rename or a lookup, medium or high for routine fixes, and xhigh when a failing test has resisted two attempts already.

  • Structured work

    Replies your code can trust

    Pass a JSON schema in response_format, or declare functions in tools and let Hydrogen ask for the next step while your application decides what actually runs.

Before you run it

What to know before you run Gemini 4 Hydrogen

Chat
This endpoint accepts text and images. Send audio, video and PDF material as extracted text or as page images.
Chat
Each reply stops at 65,536 tokens including thinking. The long-decode mode described in Google documentation is not offered here.
Chat
Google's own safety systems apply to every Hydrogen request, and some prompts may be declined.

Use cases

What to build with the Gemini 4 Hydrogen API

  • Fix a failing test

    Paste the stack trace and the two files it touches, then ask Hydrogen for the smallest patch and the reason the test broke.

  • Run an agent step by step

    Give the model your function list, execute each call it requests and return the result, keeping the step limit in your own code.

  • Read a screenshot during support

    Attach a screenshot of an error dialog and ask which setting caused it and what the user should try next.

Prompts

Gemini 4 Hydrogen prompt examples you can copy

  • Chat

    Find the smallest patch

    Here is a failing test and the function it covers. Explain the cause in two sentences, then give the smallest patch that makes the test pass.

    Why it works. A narrow request with a clear finish line suits Hydrogen and keeps thinking tokens under control.

    Try this prompt

Pricing

Gemini 4 Hydrogen API pricing

From $1.32 per million input tokens and $6.62 per million output tokens.

Get started

How to use the Gemini 4 Hydrogen API in three steps

  1. Step 1

    Begin at a lower level

    Try low or medium first and move up only when answers miss something. Higher levels spend more output tokens before the reply begins.

  2. Step 2

    Budget for thinking

    Thinking is counted inside max_tokens. A cap of a few dozen tokens can be used up before any visible text appears, so leave generous room.

  3. Step 3

    Keep the stable part first

    Put instructions and long reference text at the start of the conversation. Repeated openings can be served from the cache at the lower cached-input rate.

Parameter reference and code samples are in the API and Readme tabs at the top of this page.

FAQ

Gemini 4 Hydrogen API FAQ

About Gemini 4 Hydrogen

Choose Gemini 4 Hydrogen when the work is frequent and practical: patches, reviews, tool calls and short analysis. Start at a modest thinking level, leave room in max_tokens, and raise the level only for the problems that need it.

All Gemini models14

Text models9

Related models and next tasks