API ManyНовые модели: Qwen 3.8 Max, Claude Haiku 5.5, Claude Opus 5.5, Claude Sonnet 5.5

Google · Text & reasoning · Thinking · low · medium · high · xhigh · Tool use

Gemini 4 Hydrogen API

Call Gemini 4 Hydrogen with the same key and request shape as every other model in the catalog.

Test it onlineIntegration docs$1.875 in · $9.375 out per 1M tokens
Gemini 4 Hydrogen example

WHY IT FITS

Inside the gateway

Gemini 4 Hydrogen is Google's fast Gemini 4 model for everyday coding and agent work, and Google's recommended upgrade from Gemini 3.8 Flash. Text, images, video, audio and PDFs go in and text comes out; input, thinking and output share one 1,048,576-token window, with up to 65,536 output tokens per response by default. Thinking runs at low, medium, high (default) or xhigh, and thinking tokens are billed as output. Point any OpenAI SDK at /v1/chat/completions with your API Many key; every response carries token usage and the USD cost.

OpenAI-compatible

Call Gemini 4 Hydrogen with the standard chat completions request shape and your API Many key.

Thinking levels

Set reasoning_effort to low, medium, high, xhigh; high is the default. xhigh is for hard multi-step coding and agent tasks.

Per-token pricing

$1.8750 per million input tokens and $9.3750 per million output tokens; thinking tokens are billed as output.

One 1M window

Input, thinking and output share 1,048,576 tokens, with up to 65,536 output tokens per response by default.

Multimodal input

Send text, images, video, audio or PDFs; responses are text only.

INTEGRATION

Wire it in four steps

  1. Create a key

    Open the dashboard, mint a server-side API key and top up a small balance.

  2. Discover

    Call GET /v1/models to confirm gemini-4-hydrogen, its live specification and the per-token rate.

  3. Send

    POST /v1/chat/completions with the model id and your messages, using any OpenAI-compatible client.

  4. Read

    The reply comes back in the same response (or as a stream), with token usage and the USD cost.

CALLING

Make your first call

# OpenAI-compatible chat completions · reasoning_effort: low | medium | high | xhigh
curl https://apimany.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemini-4-hydrogen","messages":[{"role":"user","content":"Hello!"}],"reasoning_effort":"high"}'

ONLINE TEST

Send a real message from this page

The request runs against this site's own /v1/chat/completions using your signed-in balance, or an API key if you are not signed in. A pasted key stays in this browser tab and is never stored; the result is charged at the quoted USD price.

Or sign in to chat with your account balance.

GOOD TO KNOW

Integration questions

Is Gemini 4 Hydrogen OpenAI-compatible?
Yes. Send model gemini-4-hydrogen to POST /v1/chat/completions with a messages array, exactly like the OpenAI SDK. Streaming, tools and structured output follow the OpenAI request format.
How do I set the Gemini 4 Hydrogen thinking level?
Pass reasoning_effort: "low", "medium", "high", "xhigh". Without it the model thinks at "high". "minimal" is not supported and is rejected with a 400 before anything is charged.
How is Gemini 4 Hydrogen billed?
Per token: $1.8750 per million input tokens and $9.3750 per million output tokens. Thinking tokens count as output. Usage and the USD cost come back on every response, and failed requests are not charged.
What context window does Gemini 4 Hydrogen have?
Input, thinking and output share one 1,048,576-token window. A response returns at most 65,536 output tokens by default, and thinking tokens count toward max_tokens, so leave room at higher thinking levels.
What can Gemini 4 Hydrogen not do?
Gemini 4 Hydrogen returns text only: it does not generate images or audio and does not support the Live API.

KEEP EXPLORING

Nearby models