MCP / Models / Language models

Gemini 4 Hydrogen MCP: fast everyday apps with no API key

Google Gemini 4 Hydrogen is the Flash-class Gemini 4 that Google recommends in place of Gemini 3.8 Flash, with image, audio, video and PDF input. Run it from Claude Code, Codex, Cursor or any MCP client through one FlyMy.AI MCP, or call its API.

No API key$1.50 / $7.50 per 1M tokensChat modelGoogleCommercial use
Gemini 4 Hydrogen sample from the FlyMy.AI catalog

$ YOUR NEXT RUN

Claude CodeCodexCursorAny agent

Use Gemini 4 Hydrogen in your assistant

Paste this prompt into Claude Code, Codex, Cursor or any agent. It reads the guide and helps you connect.

Read https://flymy.ai/skill.md and https://flymy.ai/mcp/models/gemini-4-hydrogen.md. Help me connect the FlyMy.AI MCP to this assistant, then run Gemini 4 Hydrogen (flymyai/gemini-4-hydrogen) for this task: Ask Gemini 4 Hydrogen: plan the steps to migrate this API to a new version.
Open FlyMy.AIDownload brief

What Gemini 4 Hydrogen does

It always thinks. Send a prompt, or a chat history in messages with images as image_url parts, and get a text answer of up to 65,536 tokens from a 1M-token context. reasoning_effort is low, medium or high (medium by default); system_prompt and max_tokens work as usual, and private_mode keeps the prompt and answer out of FlyMy.AI storage and logs on synchronous calls. Call it with your FlyMy.AI key in three ways: the FlyMy.AI API, the OpenAI-compatible Chat Completions endpoint, or Google's own API at the model's FlyMy.AI address with the official Google GenAI SDK (generateContent and streamGenerateContent). On the last two your request reaches Google as you send it, so function tools, structured output, prompt caching and streaming work as Google documents them, and FlyMy.AI stores neither the prompt nor the answer. The Google API route also takes audio, video and PDF, inline or by a public URL. Billing follows the tokens Google reports, at Google's own price. Tools Google runs itself, such as Google Search and Maps grounding, are not offered.

Guide: LLM API: GPT, Claude and Gemini with One Key

Gemini 4 Hydrogen inputs

Every input of flymyai/gemini-4-hydrogen as its live schema lists them. Returns output (string), finish_reason (string), stream_details (string).

InputType, values and default
# prompt
string required What to ask Gemini 4 Hydrogen
# messages
string optional Chat history as a JSON array of {role, content}; used instead of the prompt
# system_prompt
string optional Instructions for the whole chat
# reasoning_effort
string optional Allowed: low, medium, high Default: medium How much effort Gemini 4 Hydrogen puts into the answer, reasoning included
# max_tokens
integer optional Allowed: 1 to 65536 Most tokens to generate, reasoning included
# private_mode
boolean optional Default: false Keep the prompt and the answer out of FlyMyAI storage and logs (synchronous requests only)

Ask your agent

Requests to paste as they are, once the FlyMy.AI MCP is connected.

“Ask Gemini 4 Hydrogen: plan the steps to migrate this API to a new version.”runs flymyai/gemini-4-hydrogen
“Explain this stack trace and suggest a fix with Gemini 4 Hydrogen.”runs flymyai/gemini-4-hydrogen
“With Gemini 4 Hydrogen, turn these meeting notes into a list of action items with owners.”runs flymyai/gemini-4-hydrogen

Gemini 4 Hydrogen price

$1.50 per 1M input tokens, $7.50 per 1M output tokens, cached input $0.075 per 1M. Cached input is $0.075 per 1M tokens. Billed to your FlyMy.AI balance; no seats, no subscription and no minimum.

Call Gemini 4 Hydrogen from code

The same model, from an agent or from your own code. Keys stay in your FlyMy.AI account.

From an agent, through the MCP

The agent calls get_model_schema, then run_model with:

{
  "endpoint_id": "flymyai/gemini-4-hydrogen",
  "input": {
    "prompt": "Explain in two sentences what an MCP server is."
  }
}

From your code, over HTTP

Form fields, not JSON; files go as uploads or public URLs. Text models answer on the streaming route.

curl -X POST https://api.flymy.ai/api/v1/flymyai/gemini-4-hydrogen/predict/stream/ \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -F 'prompt=Explain in two sentences what an MCP server is.'

With the OpenAI SDK

Gemini 4 Hydrogen also answers the OpenAI Chat Completions protocol.

from openai import OpenAI

client = OpenAI(base_url="https://api.flymy.ai/api/v1/flymyai/gemini-4-hydrogen/openai/v1", api_key="YOUR_FLYMYAI_KEY")
reply = client.chat.completions.create(
    model="flymyai/gemini-4-hydrogen",
    messages=[{"role": "user", "content": "Explain in two sentences what an MCP server is."}],
)
print(reply.choices[0].message.content)

Questions about Gemini 4 Hydrogen

Do I need a Google API key for Gemini 4 Hydrogen?

No. Gemini 4 Hydrogen runs on FlyMy.AI's account, so you need no Google key or plan: sign in to FlyMy.AI once and each run is billed to your FlyMy.AI balance.

How much does Gemini 4 Hydrogen cost?

$1.50 per 1M input tokens, $7.50 per 1M output tokens, cached input $0.075 per 1M. There is no subscription or minimum; the price comes from FlyMy.AI billing and was checked on 2026-11-18.

How do I use Gemini 4 Hydrogen in Claude Code, Codex or Cursor?

Add the FlyMy.AI MCP with one command and sign in, then ask in plain words, for example: "Ask Gemini 4 Hydrogen: plan the steps to migrate this API to a new version." Your agent reads the inputs with get_model_schema and runs flymyai/gemini-4-hydrogen with run_model.

What does Gemini 4 Hydrogen need as input?

Gemini 4 Hydrogen requires prompt; it also takes 5 optional inputs, each listed on this page with its allowed values and default.

Can I call Gemini 4 Hydrogen from my own code?

Yes. Send form fields to https://api.flymy.ai/api/v1/flymyai/gemini-4-hydrogen/predict/stream/ with your FlyMy.AI API key in the X-API-KEY header; it also answers the OpenAI Chat Completions protocol at https://api.flymy.ai/api/v1/flymyai/gemini-4-hydrogen/openai/v1.

Go to the source

From the FlyMy.AI model catalog of 2026-11-18: inputs from the model's live schema, prices from FlyMy.AI billing. Requests are examples to paste, not recorded runs.