Gemini 4 Hydrogen — coding and agents at your pace
Gemini 4 Hydrogen is Google's text model for everyday coding and agent workflows. Work with documents and media in a shared 1M-token context, with four thinking levels to suit the task. $0.675 / 1M input · $3.375 / 1M output.
- 1M-token context
- Long decode up to 1M
- Four thinking levels
- Function calling
- Multimodal input
Playground
Try native Gemini 4 Hydrogen requests with your account balance. Select an available price tier and review the request before sending it.
gemini-4-hydrogen$0.675 in · $3.375 out per 1M tokensMore Gemini parameters
Try one of these prompts
Messages are billed from your balance at the rates below, like an API call. Failed requests are free.
Gemini 4 Hydrogen pricing
Gemini 4 Hydrogen prices below come from current billing settings. Each available price tier appears alongside its calling ID. Thinking tokens count as output, even when the visible answer is short.
Example sizes are for illustration. Your bill follows the tokens each request reports. Current token rates. Cached input is billed once at its own rate; thinking counts as output. Check tool and processing-tier availability before using them.
| Category | Defaultgemini-4-hydrogen | google llm vertextgemini-4-hydrogen:googlellm-vertext |
|---|---|---|
| Price · 1M tokens | Price · 1M tokens | |
| Input | $0.675 | $1.05 |
| Output, including thinking | $3.375 | $5.25 |
| Cached input | $0.03375 | $0.0525 |
What is Gemini 4 Hydrogen?
Gemini 4 Hydrogen is Google DeepMind's Gemini 4 model for everyday coding, reasoning and agent workflows. It reads text, images, audio, video and PDFs, then returns text you can review, parse or pass to the next step of an application.
Gemini 4 Hydrogen uses the same native request format as Carbon and Argon. Its default thinking level is high, and the additional xhigh setting gives you another option for difficult work. Set the effort explicitly when comparing models so the defaults do not change your experiment.
Google model specification and migration guide checked November 13, 2026. Input, thinking and output share one context window.
Two ways to use Gemini 4 Hydrogen
Use Gemini 4 Hydrogen from your application or coding agent.
Call it from your backend
Use the native Gemini request format for Gemini 4 Hydrogen. The model ID belongs in the endpoint path.
- 1Create an API key
- 2Put gemini-4-hydrogen in the endpoint path
- 3Send messages in contents and settings in generationConfig
Hand it to your coding agent
Copy the prompt, then review the request and estimated cost in your agent.
- 1Copy the prompt
- 2Paste it into your agent
- 3Approve the request
What Gemini 4 Hydrogen is good at
Bring a concrete task and enough context to check the answer.
Find the real root cause
Give Gemini 4 Hydrogen the failing test, the stack trace and the code paths involved. Ask it to isolate the cause and check the fix against your tests before you merge.
Read dense documents
Gemini 4 Hydrogen reads long PDFs, charts and tables in one request. Ask for answers tied to the source page, and a structured output that follows your schema.
Connect a tool to a task
Give Gemini 4 Hydrogen your function declarations and inspect the returned arguments. Your application decides whether to execute the function and supplies its result.
Write very long outputs
With long decode, Gemini 4 Hydrogen can keep going past 64K tokens, up to 1M minus your input, by continuing from where the last response stopped.
Practical Gemini 4 Hydrogen tasks
These are prompts to try with your own evaluation data. No measured quality or speed advantage is claimed.
Plan a multi-package migration
Share the dependency graph and failing builds. Gemini 4 Hydrogen can order the steps and flag the packages that break first.
Extract a document into JSON
Give Gemini 4 Hydrogen an invoice or report and a response schema. Check missing fields and source references before writing the result into your application.
Explain a recorded workflow
Supply a short video and ask Gemini 4 Hydrogen to identify the steps, timestamps and ambiguous actions. Review its observations against the recording.
How to use Gemini 4 Hydrogen
- 01
Prepare your account
Sign in, add credit and create an API key for server-side use. Open Gemini 4 Hydrogen to see the available price tiers.
- 02
Assemble the conversation
Put user messages, earlier model parts and tool results in contents. Give Gemini 4 Hydrogen the files it needs with correct MIME types.
- 03
Set thinking and output
Gemini 4 Hydrogen starts at high thinking. Choose low, medium, high or xhigh deliberately, and leave enough output budget for both reasoning and the answer.
- 04
Handle the result
Read Gemini 4 Hydrogen candidates, finishReason and usageMetadata. Keep full content parts and thought signatures when the conversation continues.
The native Gemini 4 Hydrogen API
Gemini 4 Hydrogen uses the native generateContent and streamGenerateContent endpoints. Switching from Gemini 4 Argon needs no request changes besides the model ID.
Use the stable identifier
The URL contains gemini-4-hydrogen. Gemini 4 Hydrogen needs no preview suffix and no separate name for streaming.
Leave sampling out
Gemini 4 Hydrogen ignores temperature, topP and topK, and explicit context caching is not supported. Implicit caching applies on its own from 4,096 prompt tokens.
Continue long outputs
A response that ends with finishReason CONTINUATION carries a continuationToken. Send the same Gemini 4 Hydrogen request again with that token to get the next part.
Gemini 4 Hydrogen vs Gemini 4 Argon
Gemini 4 Hydrogen and Argon share a context window and request format. Their thinking defaults differ, so compare them with the same explicit settings before selecting one for your application.
| Setting | Gemini 4 Hydrogen | Gemini 4 Argon |
|---|---|---|
| Default thinking | high | medium |
| Thinking levels | low, medium, high, xhigh | low, medium, high |
| Shared context | 1,048,576 tokens | 1,048,576 tokens |
| Non-streaming REST output per hop | 65,536 tokens | 32,768 tokens |
| Request format | Native Gemini | Native Gemini |
Use the same input, response format and thinking level for a comparison. A larger budget alone does not establish a better answer. Google Gemini 4 Hydrogen guide
Gemini 4 Hydrogen limits
One shared window
Input, thinking and output share 1,048,576 tokens. Each Gemini 4 Hydrogen response can return up to 65,536 tokens, with or without streaming. Longer answers require continuation requests.
Text out, media in
Gemini 4 Hydrogen reads images, audio, video and PDFs and returns text. It does not generate images or audio, and it has no Live API.
Budget for the full answer
Thinking uses part of the output budget. A low cap can stop Gemini 4 Hydrogen before it produces useful text. Inspect the finish reason before deciding to send another request.
Verify advanced tools
Google lists Search, Maps, code execution, URL context, file search and computer use for Gemini 4 Hydrogen. Availability through your account still needs checking for each feature.
Use Gemini 4 Hydrogen with SeedRouter
One balance across models
Pay for Gemini 4 Hydrogen from the same SeedRouter balance as the other listed models. Credit never expires.
Current rates in context
See Gemini 4 Hydrogen prices beside the model settings. The Playground estimates token cost with your selected group's rate.
Failed requests are free
Failed Gemini 4 Hydrogen requests are not charged. On success, usage shows input, cached input, output and thinking separately.
Explore other text models
Compare available text models by their own documented capabilities and current rates.
Call GPT-6 Astra with the OpenAI SDK. OpenAI's most capable model, with a 1.05M-token context, live per-token pricing and a playground you can try right now.
View pricingCall GPT-6 Sol with the OpenAI SDK. 1.05M-token context, reasoning effort from none to max, live per-token pricing and a playground you can try right now.
View pricingCall GPT-6.1 Sol with the OpenAI SDK: near-Astra coding and agent work at a fifth of the price, 1.05M-token context, 95% off cached input. Try it now.
View pricingCall GPT-6 Luna with the OpenAI SDK. The lowest-cost GPT-6 model, with a 1.05M-token context, live per-token pricing and a playground you can try right now.
View pricingTry Claude Fable 5.2 for coding and problem solving with a 1M-token context. Compare live token prices and model tiers, then test it in the playground.
View pricingTry Claude Mythos 5 for complex reasoning and research with a 1M-token context. Compare live token prices and model tiers, then test it in the playground.
View pricingGemini 4 Hydrogen guides
Notes on Gemini 4 Hydrogen settings and choosing between models.
Gemini 4 Hydrogen FAQ
What is Gemini 4 Hydrogen?+
Gemini 4 Hydrogen is a Google DeepMind text model for coding, reasoning and agent workflows. It accepts text and media in a shared 1,048,576-token context. Its output is text; reading a picture does not mean it can generate one.
Which Gemini 4 Hydrogen model ID should I use?+
Use the stable ID gemini-4-hydrogen in the endpoint path. Set the thinking level in generationConfig rather than adding a suffix to the model name.
Should I use Gemini 4 Hydrogen or Gemini 4 Argon?+
Gemini 4 Hydrogen adds xhigh thinking and defaults to high, while Argon defaults to medium. Start by running a task with a known answer on both at the same explicit thinking level. Compare correctness, usage and elapsed time on your workload.
What thinking level does Gemini 4 Hydrogen use by default?+
Gemini 4 Hydrogen defaults to high and supports low, medium, high and xhigh. Thinking cannot be turned off. Use the lowest level that passes your checks, then increase it when the task needs more reasoning.
How long can a Gemini 4 Hydrogen answer be?+
A Gemini 4 Hydrogen hop returns up to 65,536 tokens in both JSON and streaming modes. Continuation requests can extend total output to the shared context limit minus input. Each additional request is separately billed; it does not run automatically in the Playground.
How is Gemini 4 Hydrogen billed?+
Gemini 4 Hydrogen bills input, cached input and output tokens. Thinking is part of output. The price table above reads current SeedRouter rates; it does not copy a fixed dollar amount from a launch announcement. Failed requests are not charged.
How do Gemini 4 Hydrogen price tiers work?+
Default uses gemini-4-hydrogen without a suffix. Additional tiers use the complete model:group ID shown beside their price. The Playground lists only tiers available to your account; an explicit choice stays within that tier.
Try Gemini 4 Hydrogen
Bring a coding task or document you can check. Try Gemini 4 Hydrogen, inspect its answer and token usage, and keep the settings that work for you.