> Markdown mirror of https://router.one/blog/gemini-4-hydrogen-api-guide for AI assistants and crawlers. Router One is a unified, OpenAI-compatible LLM API gateway.
> Published: 2026-11-13 · Author: Router One Team

# Gemini 4 Hydrogen API Guide: Model ID, OpenAI SDK, Billing

_Gemini 4 Hydrogen on Router One: the google/gemini-4-hydrogen catalog ID, OpenAI-compatible Chat Completions, wallet billing, and how it differs from Carbon._

As of the 2026-11-13 catalog, Router One lists Gemini 4 Hydrogen as `google/gemini-4-hydrogen`. Send `"model": "google/gemini-4-hydrogen"` to `https://api.router.one/v1/chat/completions` with a Router One key — or point any OpenAI-compatible SDK or client at `https://api.router.one/v1` — and the request is routed, metered and logged like every other model. As of the 2026-11-13 plan response, no tier of the Pro, Max or Ultra plans lists Gemini 4 Hydrogen, so every call bills per token to the wallet at the rates on the [Gemini 4 Hydrogen model page](https://router.one/models/gemini-4-hydrogen), plan or no plan.

This guide covers what the catalog lists for the id, what Google documents about the model, how billing works, how it differs from Gemini 4 Carbon, Gemini 4 Argon and Gemini 3.8 Flash, the first request with curl and the OpenAI SDKs, which clients can call it, and access from mainland China; every Google statement below is attributed and checked 2026-11-13, and the catalog and plan observations are dated. The live catalog is the source of truth for a new request.

## Gemini 4 Hydrogen at a glance

| Field | What the catalog lists (2026-11-13) |
| --- | --- |
| Catalog id | `google/gemini-4-hydrogen` |
| Input / output | text and image in, text out |
| Capability flags | chat, streaming, tool calling, vision |
| Endpoint | `POST /v1/chat/completions` (OpenAI-compatible) |
| Not served on | any other model endpoint, including `POST /v1/messages` (Claude-family and DeepSeek ids only) |
| Plans | in no plan tier of Pro, Max or Ultra (2026-11-13 plan response): every call bills the wallet |
| Rates | per token, posted on the [model page](https://router.one/models/gemini-4-hydrogen); this guide prints no figures |

## What Google documents for Gemini 4 Hydrogen

Per Google's [Gemini API release notes](https://aistudio.google.com/docs/changelog) (November 12, 2026 entry; checked 2026-11-13), Gemini 4 Hydrogen (Google's model id `gemini-4-hydrogen`) is generally available on the Gemini API. Google's [Gemini 4 Hydrogen model page](https://aistudio.google.com/docs/models/gemini-4-hydrogen) (checked 2026-11-13) describes it as bringing "cutting-edge intelligence to everyone, with speed and efficiency for everyday coding and agentic workflows."

Google's Gemini 4 Hydrogen model page (checked 2026-11-13) lists the following for Google's own Gemini API:

- **Status.** Google lists the model as a stable version, and Google's [deprecations page](https://aistudio.google.com/docs/deprecations) (checked 2026-11-13) gives November 12, 2026 as its release date and no announced shutdown date.
- **Token limits.** A context window of 1,048,576 tokens that input and output share, the unified 1M-token context window that Google's [What's new in Gemini 4 Hydrogen](https://aistudio.google.com/docs/whats-new-gemini-4-hydrogen) page describes (checked 2026-11-13), with a maximum output of 65,536 tokens by default. Google's long decode feature, which runs on Google's own APIs, lets a response go past 65,536 tokens up to the rest of that window.
- **Inputs and output.** Per Google, text, image, video, audio and PDF in, text out. The Router One catalog entry lists text and image input, so the samples below send only those.
- **Features.** Google lists function calling, structured outputs, code execution, search grounding, URL context and implicit context caching as supported (explicit caching is not), and image generation, audio generation and the Live API as not supported.
- **Thinking.** Google lists the levels `low`, `medium`, `high` and `xhigh`. The next section has Google's default and the details.

Those are Google's statements about its own Gemini API; Router One is responsible for the catalog entry and for how the gateway routes and bills it. The catalog lists the tool-calling flag for this id, so function tools go in the standard Chat Completions `tools` array. Google's built-in tools are Gemini API features: on Router One's Chat Completions endpoint the hosted search tool is handled for `google/gemini-3-flash` only, and for every other id the tools array is forwarded as sent ([API compatibility facts](https://router.one/facts/api-compatibility.md)). Test the tool definitions your code sends on your own prompts before you move a workload.

## Thinking levels and reasoning effort

Per Google's What's new in Gemini 4 Hydrogen page and its [thinking guide](https://aistudio.google.com/docs/thinking) (both checked 2026-11-13):

- **Levels.** Gemini 4 Hydrogen thinks at `low`, `medium`, `high` or `xhigh`, and `high` is the default; Google recommends starting with `high` for most complex tasks, keeping `xhigh` for hard, multi-step problems where `high` falls short, and running your evaluations at all four levels before picking the lowest one that meets your quality bar.
- **Token use.** Per Google, `xhigh` raises latency and token use and can overthink simpler prompts; to use fewer tokens, Google suggests lowering the thinking level rather than setting a small output cap.
- **Output cap.** Google's output cap counts thinking tokens as well as the answer, so a response that reaches it while the model is still thinking comes back truncated or empty; for standard tasks Google recommends keeping the 65,536-token default, and raising hard-coded low caps to it.
- **No off switch.** Google's [OpenAI compatibility guide](https://aistudio.google.com/docs/openai) (checked 2026-11-13) maps a `reasoning_effort` of `minimal` or `low` to the `low` thinking level for Gemini 4 models, `medium` to `medium`, `high` to `high` and, for Gemini 4 Hydrogen only, `xhigh` to `xhigh`; it uses the model's default level when the field is absent, and says reasoning cannot be turned off for Gemini 4 models.

Those rules describe Google's own API. The samples below do not set `reasoning_effort`; if your code sets one, run it on your own prompts before you move a workload. Thinking output is billed at the model's posted output rate, with no separate reasoning line ([pricing facts](https://router.one/facts/pricing.md)), so thinking tokens add to a call's wallet charge.

## How billing works

Gemini 4 Hydrogen bills per token at the input and output rates posted on its [model page](https://router.one/models/gemini-4-hydrogen). Router One posts its own rates for each model, with pricing up to 90% off official prices on select models, and this guide prints no per-token figures because they go stale. As of the 2026-11-13 plan response, no tier of the Pro, Max or Ultra plans lists Gemini 4 Hydrogen, so every call bills the wallet at the posted rates, plan or no plan.

Google prices its own Gemini API separately: per Google's [pricing page](https://aistudio.google.com/docs/pricing) and its What's new in Gemini 4 Hydrogen page (both checked 2026-11-13), Gemini 4 Hydrogen is offered on Google's free tier, Google lists one flat standard price for it across the full 1M-token context window, and Google's output price includes thinking tokens. That price list is Google's; the rate Router One charges for the id is the one on its model page.

Dashboard → Logs shows each call's model, tokens, wallet charge or plan quota used, latency and status ([per-request observability](https://router.one/llm-observability)), so you can add up a task's wallet charges from the calls it made rather than estimate them from rates.

## Do subscription plans cover Gemini 4 Hydrogen?

No, as of the 2026-11-13 plan response: no tier of Router One's Pro, Max or Ultra plans lists Gemini 4 Hydrogen, so on a plan its calls bill the wallet per token at the posted rates, as they do without one. Gemini 4 Carbon and Gemini 4 Argon are separate catalog entries: as of the 2026-11-13 plan response, both are in the Premium models tier of the Pro, Max and Ultra plans. Plan model lists change; the [pricing page](https://router.one/pricing) lists each plan's allowance and the live model lists.

## Gemini 4 Hydrogen, Gemini 4 Carbon or Gemini 4 Argon?

As of the 2026-11-13 catalog, Router One also lists [Gemini 4 Carbon](https://router.one/models/gemini-4-carbon) (`google/gemini-4-carbon`) and [Gemini 4 Argon](https://router.one/models/gemini-4-argon) (`google/gemini-4-argon`). Per Google's What's new in Gemini 4 Hydrogen page (checked 2026-11-13), the three models share the same unified 1M-token context window, long decode architecture and built-in tools on Google's own Gemini API, and Google positions them this way:

- **Gemini 4 Hydrogen.** Fast and efficient on complex agentic workflows; Google suggests it for everyday coding, interactive web apps, tool-using agents and multimodal document workflows, and calls it its recommended upgrade from Gemini 3.8 Flash and earlier Gemini 3 and 2.5 models.
- **Gemini 4 Carbon.** Google's strongest frontier model, which spends more thinking tokens per turn to test hypotheses and verify edge cases on hard architecture, debugging, science and security audits.
- **Gemini 4 Argon.** For most advanced long-horizon tasks, with lower token use than Gemini 4 Carbon.

Per the same Google page (checked 2026-11-13), adding or switching to Gemini 4 Hydrogen from Gemini 4 Carbon or Gemini 4 Argon needs no request schema changes on Google's API; Gemini 4 Hydrogen defaults to the `high` thinking level and supports `xhigh`, while Gemini 4 Carbon and Gemini 4 Argon default to `medium` and cap at `high`.

What Router One controls:

- **Endpoint.** All three are served on `/v1/chat/completions` only, so the key, base URL and endpoint stay the same: on the wire, the switch is the model string.
- **Plans.** As of the 2026-11-13 plan response, Gemini 4 Carbon and Gemini 4 Argon are in the Premium models tier of the Pro, Max and Ultra plans, so on a plan their calls draw that tier's allowance. As of the 2026-11-13 plan response, no tier lists Gemini 4 Hydrogen, so its calls bill the wallet per token, plan or no plan.
- **Posted rates.** Each model page posts its own per-token rates; compare the [Gemini 4 Hydrogen](https://router.one/models/gemini-4-hydrogen), [Gemini 4 Carbon](https://router.one/models/gemini-4-carbon) and [Gemini 4 Argon](https://router.one/models/gemini-4-argon) pages, and run your own prompts on each before you move a workload.

The [Gemini 4 Carbon API guide](https://router.one/blog/gemini-4-carbon-api-guide) and the [Gemini 4 Argon API guide](https://router.one/blog/gemini-4-argon-api-guide) cover those two ids, and the [Gemini 4 Carbon vs Gemini 4 Argon](https://router.one/models/compare/gemini-4-carbon-vs-gemini-4-argon) comparison page renders their spec sheets and live rates side by side.

## Moving from Gemini 3.8 Flash

Per the migration checklist on Google's What's new in Gemini 4 Hydrogen page (checked 2026-11-13), these are the changes from Gemini 3.8 Flash on Google's own API:

- **Context.** Gemini 3.8 Flash keeps separate input and output budgets, while Gemini 4 Hydrogen shares one 1,048,576-token window across input, thinking and output with a 65,536-token default output cap; Google says to remove hard-coded low output caps or raise them to 65,536.
- **Thinking.** Gemini 4 Hydrogen defaults to `high` rather than `medium`, adds `xhigh`, and rejects a `minimal` thinking level with an error; Google suggests testing `medium` and `high` on your evaluation suite.
- **Caching.** Gemini 4 Hydrogen does not support explicit context caching; implicit caching runs by default.
- **Sampling parameters.** Remove `temperature`, `top_p` and `top_k`; Google's server ignores custom values.

On Router One, both ids are served on `/v1/chat/completions` with the same key and base URL, so the request change is the model string. As of the 2026-11-13 plan response, Gemini 3.8 Flash is in the Standard models tier of the Pro, Max and Ultra plans. As of the 2026-11-13 plan response, no tier lists Gemini 4 Hydrogen, so its calls bill the wallet per token, plan or no plan. The [Gemini 3.8 Flash API guide](https://router.one/blog/gemini-3-8-flash-api-guide) covers that id.

## Send the first request

1. **Create a key.** Dashboard → API Keys → Create Key. Keys look like `sk-...`. For a trial, give the key a `maxSpend` cap. A key's spend cap (maxSpend) limits what that key can charge to your wallet: once its cumulative wallet charges reach the cap, its wallet-billed requests return HTTP 402. Requests covered by your plan's quota draw from the account's shared plan quota and do not count toward any key's cap. Your other keys keep working ([per-key cost tracking](https://router.one/llm-cost-tracking)).
2. **Keep a wallet balance.** Gemini 4 Hydrogen calls bill the wallet, plan or no plan: no tier of the 2026-11-13 plan response lists it — top up in Dashboard → Deposit.
3. **Set the base URL** to `https://api.router.one/v1` in any OpenAI-compatible SDK or client.
4. **Call the model.** The examples use the catalog id `google/gemini-4-hydrogen` and set `max_tokens` explicitly.

Chat Completions with curl, streaming:

```bash
curl https://api.router.one/v1/chat/completions \
  -H "Authorization: Bearer sk-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-4-hydrogen",
    "max_tokens": 65536,
    "stream": true,
    "messages": [
      {"role": "system", "content": "You review pull requests. List blocking issues first, then suggestions."},
      {"role": "user", "content": "A handler loops over 500 orders and runs a separate query to load the customer for each order. What is wrong, and how should the query change?"}
    ]
  }'
```

The same request without streaming, from the OpenAI Python SDK (`pip install -U openai`) — only `base_url` and the key change:

```python
from openai import OpenAI

client = OpenAI(
    base_url="https://api.router.one/v1",
    api_key="sk-your-api-key",
)

response = client.chat.completions.create(
    model="google/gemini-4-hydrogen",
    max_tokens=65536,
    messages=[
        {"role": "system", "content": "You review pull requests. List blocking issues first, then suggestions."},
        {"role": "user", "content": "A handler loops over 500 orders and runs a separate query to load the customer for each order. What is wrong, and how should the query change?"},
    ],
)
print(response.choices[0].message.content)
print(response.choices[0].finish_reason, response.usage)
```

The reply has the Chat Completions shape: the text is in `choices[0].message.content`, `finish_reason` says why generation stopped, and `usage` carries the input and output token counts.

Streaming from Node with the OpenAI SDK (`npm install openai`):

```ts
import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.router.one/v1",
  apiKey: "sk-your-api-key",
});

const stream = await client.chat.completions.create({
  model: "google/gemini-4-hydrogen",
  max_tokens: 65536,
  stream: true,
  messages: [
    {
      role: "user",
      content:
        "A handler loops over 500 orders and runs a separate query to load the customer for each order. Explain the problem in three sentences.",
    },
  ],
});

for await (const chunk of stream) {
  process.stdout.write(chunk.choices[0]?.delta?.content ?? "");
}
```

An image, sent inline as a base64 data URL (Python, same client as above):

```python
import base64

with open("settings-page.png", "rb") as f:
    image_b64 = base64.b64encode(f.read()).decode()

response = client.chat.completions.create(
    model="google/gemini-4-hydrogen",
    max_tokens=65536,
    messages=[{
        "role": "user",
        "content": [
            {"type": "text", "text": "List the layout and accessibility bugs in this screenshot of a settings page, numbered."},
            {"type": "image_url", "image_url": {"url": f"data:image/png;base64,{image_b64}"}},
        ],
    }],
)
print(response.choices[0].message.content)
```

Rules that hold for every request:

- **Set `max_tokens` on every request.** Leave room above the longest answer you expect; the samples above set 65536, the default output cap Google recommends keeping for standard tasks on this model (What's new in Gemini 4 Hydrogen, checked 2026-11-13). On Router One's Chat Completions, `max_completion_tokens` is accepted as the same cap (if both are sent, `max_completion_tokens` wins). Per Google's thinking guide (checked 2026-11-13), thinking tokens count toward the output cap, so leave headroom.
- **Leave out sampling parameters.** Per the migration checklist on Google's What's new page (checked 2026-11-13), remove `temperature`, `top_p` and `top_k`; on Google's own Gemini API the server ignores custom values. Put tone and format rules in the system message instead.
- **Stream long answers.** Send long answers with `stream: true` and give your HTTP client a generous timeout ([streaming guide](https://router.one/llm-streaming)).
- **Keep a healthy wallet balance.** Before a request runs, the gateway reserves an estimate against your balance; when the balance cannot cover it, the gateway lowers `max_tokens` to fit, which can cut a long answer short. If it cannot cover even the request's initial reservation, the gateway returns HTTP 402 before the model runs.
- **Send images inline.** Pass images as base64 data URLs in `image_url` content parts, as in the example above; Router One serves no Files API.
- **Test what your code depends on.** Before you move a workload, run the parameters and request shapes your code sends (tool definitions, image inputs, output caps, any thinking setting) on your own prompts first.

5. **Read the trace.** Dashboard → Logs shows each call's model, tokens, wallet charge or plan quota used, latency and status ([per-request observability](https://router.one/llm-observability)). Retryable upstream failures are absorbed by [automatic fallback](https://router.one/llm-fallback) across the candidate routes for the same model; a Gemini 4 Hydrogen request is never answered by a different model.

A request the gateway refuses before any model is called has no row in Dashboard → Logs; its status code and error message say why. A 401 means the key was not accepted, a 402 that the wallet balance or the key's `maxSpend` cannot cover the request's initial reservation, and a 404 that the path or the model id does not exist, so copy the id exactly as the catalog lists it. The [error codes page](https://router.one/llm-api-error-codes) covers the rest.

## Which clients can call it

Any client that speaks Chat Completions and takes a custom base URL can call Gemini 4 Hydrogen: the official OpenAI SDKs ([OpenAI SDK setup](https://router.one/integrations/openai-sdk)) and the chat apps and frameworks in the [integrations directory](https://router.one/integrations) that send `/v1/chat/completions` requests. Give the client the base URL `https://api.router.one/v1` and a Router One key, and enter `google/gemini-4-hydrogen` as the model; the [OpenAI-compatible API page](https://router.one/openai-compatible-api) explains the endpoint in general.

Claude Code cannot call Gemini 4 Hydrogen. Claude Code sends Anthropic Messages requests to `/v1/messages`, which Router One serves for Claude-family and DeepSeek ids only; for any other chat id the catalog lists, the gateway returns HTTP 400 before any model is called. Call Gemini 4 Hydrogen from a Chat Completions client. The [API compatibility facts](https://router.one/facts/api-compatibility.md) state the endpoint rule for each model family.

## From mainland China

Requests reach `api.router.one` from mainland China without a VPN, on the same key and base URL. Top up with a card or Alipay through one hosted checkout, or with USDT/USDC on six chains (Tron, BSC, Ethereum, Polygon, Base, Arbitrum). No US credit card required. For the other Gemini models Router One lists, see the [Gemini API in China page](https://router.one/gemini-api-china) and the [Gemini API in China guide](https://router.one/blog/gemini-api-china-guide).

## FAQ

**What is the model ID for Gemini 4 Hydrogen on Router One?**
The catalog id is google/gemini-4-hydrogen, as of the 2026-11-13 catalog; use it in code and client settings, copied exactly as the catalog lists it. As of the 2026-11-13 plan response, no tier of the Pro, Max or Ultra plans lists it, so its calls bill per token to the wallet.

**How much does the Gemini 4 Hydrogen API cost?**
On Router One, per token at the input and output rates posted on the Gemini 4 Hydrogen model page, with pricing up to 90% off official prices on select models; this guide prints no figures because they go stale. As of the 2026-11-13 plan response, it is in no plan tier, so every call bills the wallet. On Google's own Gemini API, per Google's pricing page (checked 2026-11-13), Google lists one flat standard price for it across the full 1M-token context window, and its output price includes thinking tokens.

**Is there a free tier for Gemini 4 Hydrogen?**
On Google's own Gemini API, yes: per Google's pricing page and What's new in Gemini 4 Hydrogen page (checked 2026-11-13), Gemini 4 Hydrogen is offered on Google's rate-limited free tier. On Router One, every Gemini 4 Hydrogen call bills the wallet at the posted rates; as of the 2026-11-13 plan response, it is in no plan tier.

**Do I need a Google API key to call Gemini 4 Hydrogen?**
Not through Router One: one Router One key (sk-...) on `https://api.router.one/v1` calls Gemini 4 Hydrogen and the other catalog models, with no Google API key and no Google Cloud billing account.

**What is the context window of Gemini 4 Hydrogen?**
Per Google's model page and What's new in Gemini 4 Hydrogen page (checked 2026-11-13), Gemini 4 Hydrogen has a context window of 1,048,576 tokens that input, thinking and output share — the unified 1M-token window Google describes — with a default maximum output of 65,536 tokens on Google's own Gemini API. The Router One model page shows the live catalog entry for the id.

**Which thinking levels does Gemini 4 Hydrogen support?**
On Google's own Gemini API, per Google (checked 2026-11-13): low, medium, high and xhigh, with high as the default; reasoning cannot be turned off for Gemini 4 models, and, per Google's OpenAI compatibility guide, reasoning_effort sets the level on Google's OpenAI-compatible endpoint. The samples in this guide do not set reasoning_effort; test any setting your code sends on your own prompts first.

**Should I use Gemini 4 Hydrogen, Gemini 4 Carbon or Gemini 4 Argon?**
Per Google's What's new in Gemini 4 Hydrogen page (checked 2026-11-13), Gemini 4 Hydrogen is fast and efficient for everyday coding and agentic workflows, Gemini 4 Carbon is Google's strongest frontier model and spends more thinking tokens per turn, and Gemini 4 Argon handles most advanced long-horizon tasks with lower token use than Gemini 4 Carbon. On Router One all three are served on /v1/chat/completions, so switching is a model-string change. As of the 2026-11-13 plan response, Gemini 4 Carbon and Gemini 4 Argon are in the Premium models tier of the Pro, Max and Ultra plans. As of the 2026-11-13 plan response, no tier lists Gemini 4 Hydrogen.

**Is Gemini 4 Hydrogen included in Router One subscription plans?**
No, as of the 2026-11-13 plan response: no tier of the Pro, Max or Ultra plans lists it, so its calls bill the wallet per token at the posted rates, plan or no plan. Gemini 4 Carbon and Gemini 4 Argon, separate catalog entries, are in the Premium models tier of those plans as of the 2026-11-13 plan response. The pricing page lists each plan's allowance.

**Can I use Gemini 4 Hydrogen in Claude Code?**
No. Claude Code sends Anthropic Messages requests to /v1/messages, which Router One serves for Claude-family and DeepSeek ids only, so a Gemini 4 Hydrogen request there returns HTTP 400 before any model is called. Call Gemini 4 Hydrogen from a Chat Completions client.

**Can I call Gemini 4 Hydrogen from mainland China?**
Yes: api.router.one is reachable from mainland China without a VPN, on the same key and base URL. Top up with a card or Alipay through one hosted checkout, or with USDT/USDC on six chains (Tron, BSC, Ethereum, Polygon, Base, Arbitrum). No US credit card required.

## Next steps

- Open the [Gemini 4 Hydrogen model page](https://router.one/models/gemini-4-hydrogen) for the live rates and the endpoint.
- Weighing Gemini 4 Carbon or Gemini 4 Argon? Read the [Gemini 4 Carbon API guide](https://router.one/blog/gemini-4-carbon-api-guide) and the [Gemini 4 Argon API guide](https://router.one/blog/gemini-4-argon-api-guide), and put those two side by side on the [Gemini 4 Carbon vs Gemini 4 Argon](https://router.one/models/compare/gemini-4-carbon-vs-gemini-4-argon) comparison page.
- Check each plan's allowance and the models each tier lists on the [pricing page](https://router.one/pricing).
- Calling other Gemini models from mainland China? See the [Gemini API in China page](https://router.one/gemini-api-china) and the [Gemini API in China guide](https://router.one/blog/gemini-api-china-guide).
- New to OpenAI-compatible endpoints? The [OpenAI-compatible API page](https://router.one/openai-compatible-api) covers base URLs, keys and SDKs.
- Moving from Gemini 3.8 Flash? Read the [Gemini 3.8 Flash API guide](https://router.one/blog/gemini-3-8-flash-api-guide).

## See also

- Canonical page: https://router.one/blog/gemini-4-hydrogen-api-guide
- Gemini API in China: https://router.one/gemini-api-china
- LLM API Gateway and Routing: https://router.one/llm-api-gateway
- All blog posts: https://router.one/blog
- Models and per-model token rates: https://router.one/models (markdown: https://router.one/models.md)
- Pricing: https://router.one/pricing
- API docs (markdown): https://router.one/docs.md
- Company facts: https://router.one/facts/company.md