Independent guide. Not affiliated with, endorsed by or sponsored by Google, Google DeepMind or Gemini. About this site
gemini4argon.comUnofficial guide

Developers

Gemini 4 Argon API: model ID, pricing and code that works today

Updated · Independent coverage, not affiliated with Google

The Gemini 4 Argon API is open. Since October 13, 2026, any developer with a paid Gemini API account can call it. Here is the model ID, what it costs, what it can do, and code you can copy and run right now.

Is the Gemini 4 Argon API available?

Yes. Google made Gemini 4 Argon generally available on October 13, 2026, in the Gemini API (paid tier), Google AI Studio, Android Studio and Vertex AI, which Google now calls Gemini Enterprise Agent Platform. It is not on the free tier.

Argon was announced on September 30 and went to trusted cyber defenders first. Thirteen days later Google opened it to every developer and to Google AI Pro and Ultra subscribers in the Gemini app. See the release date page for the full timeline.

Newer model alert. On October 28, Google announced Gemini 4 Carbon, which beats Argon on every test in Google's table at the same price. It is limited to cyber defenders for now, so Argon is still the Gemini 4 model you can actually use.

What is the Gemini 4 Argon model ID?

The model ID is gemini-4-argon. It is listed on the Gemini API models page as a stable model. Google also runs a gemini-4-argon-cyber-permissive version without the usual cyber guardrails, but only for approved members of its Fairwind security program.

To check what your own key can use, ask the API:

# pip install google-genai
from google import genai

client = genai.Client()  # reads GEMINI_API_KEY
for m in client.models.list():
    if "gemini-4" in m.name:
        print(m.name)

Or with curl:

curl -s "https://generativelanguage.googleapis.com/v1beta/models" \
  -H "x-goog-api-key: $GEMINI_API_KEY" | grep -i "gemini-4"

If gemini-4-argon doesn't show up, check that billing is turned on for your project: Argon is paid tier only.

Gemini 4 Argon API pricing

Argon launched with an introductory price that ends on November 30, 2026:

Per 1M tokensUntil Nov 30, 2026From Dec 1, 2026
Input$2.00$4.00
Output (including thinking)$10.00$20.00
Cached input$0.10$0.20
Batch or Flex (input / output)$1.00 / $5.00$2.00 / $10.00

Prices don't go up for very long prompts. Caching is automatic (implicit), so you get the cached rate whenever you resend the same opening content. Try our Gemini 4 Argon cost calculator to see what your own jobs cost before and after December 1.

Watch the output bill. With long decoding turned on, Argon can write up to about 1 million tokens in one answer. That single answer would cost about $10 now and $20 from December 1. Set a sensible output cap on every call (examples below).

What the Gemini 4 Argon API can do

Context1,048,576 tokensInput and output share it
Output limit65,536 by defaultUp to the full 1M with long decoding
InputsText, images, video, audio, PDFIncluding charts and visual documents
OutputTextNo image, audio or video generation

Argon thinks before it answers, with three thinking levels: low, medium and high. It does not support the Live API. Google hasn't published a knowledge cutoff on the model page.

Code you can run today

Keep the model name in one environment variable. That way you can try Argon against other models, and moving to Gemini 4 Carbon when it opens is a one-line change.

  1. Get an API key. Sign in to Google AI Studio, create a Gemini API key and turn on billing (Argon is paid tier only). Keep the key out of your code and out of git.
  2. Set two environment variables. GEMINI_API_KEY for the key, and GEMINI_MODEL set to gemini-4-argon.
  3. Install the official SDK. pip install google-genai for Python, or npm install @google/genai for JavaScript.
  4. Cap your output. Set a maximum output length on every call so a runaway answer can't run up the bill.

Python

import os
from google import genai
from google.genai import types

client = genai.Client()  # reads GEMINI_API_KEY
MODEL = os.environ.get("GEMINI_MODEL", "gemini-4-argon")

response = client.models.generate_content(
    model=MODEL,
    contents="Review this function and list every bug you find.",
    config=types.GenerateContentConfig(max_output_tokens=8000),  # cap the bill
)
print(response.text)

JavaScript / Node

import { GoogleGenAI } from "@google/genai";

const ai = new GoogleGenAI({}); // reads GEMINI_API_KEY
const MODEL = process.env.GEMINI_MODEL ?? "gemini-4-argon";

const response = await ai.models.generateContent({
  model: MODEL,
  contents: "Summarize this contract in plain English.",
  config: { maxOutputTokens: 8000 },
});
console.log(response.text);

curl (REST)

curl "https://generativelanguage.googleapis.com/v1beta/models/${GEMINI_MODEL:-gemini-4-argon}:generateContent" \
  -H "x-goog-api-key: $GEMINI_API_KEY" \
  -H "Content-Type: application/json" \
  -X POST \
  -d '{"contents":[{"parts":[{"text":"Explain the CAP theorem in three sentences."}]}]}'

OpenAI-compatible endpoint

Already using the OpenAI SDK? The Gemini API has an OpenAI-compatible endpoint, so you only change the key, the base URL and the model name:

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["GEMINI_API_KEY"],
    base_url="https://generativelanguage.googleapis.com/v1beta/openai/",
)
reply = client.chat.completions.create(
    model=os.environ.get("GEMINI_MODEL", "gemini-4-argon"),
    messages=[{"role": "user", "content": "Write a haiku about argon."}],
)
print(reply.choices[0].message.content)

Google's Argon announcement doesn't mention this endpoint specifically, so test your exact use case before you rely on it.

Using the long output well

Argon's headline feature for developers is how much it can write in one go. A few habits make that useful instead of expensive:

  • Stream long answers. Don't wait minutes for one giant response. Print it as it arrives:
for chunk in client.models.generate_content_stream(model=MODEL, contents=long_task):
    print(chunk.text or "", end="", flush=True)
  • Ask for a plan first. Have Argon outline the full job, check the plan, then ask it to write everything in one pass.
  • Reuse the same opening. Caching is automatic, so keep the shared part of your prompt (the codebase, the manual) at the start and identical between calls.
  • Cap every call. Set max_output_tokens to what the job needs, and raise it only for the runs that truly need a huge answer.

Building something creative?

Try Google's Veo 3.1, Gemini Omni and Nano Banana 2 side by side on Lumeta before you wire up a single API.

Open Lumeta

Google AI Studio, Vertex AI and other platforms

Argon is available to developers in the Gemini API through Google AI Studio and Android Studio, and to businesses on Vertex AI (Gemini Enterprise Agent Platform), Gemini for business and Google Antigravity. In the Gemini app and AI Mode in Search, it needs a Google AI Pro or Ultra subscription. Third-party routers and coding tools add new Gemini models on their own schedules, so check each one.

Want to use Argon from inside Claude Code or Claude Desktop? Our Gemini 4 Argon and Claude guide shows a small MCP server that does it.

Gemini 4 Argon API questions

What is the Gemini 4 Argon API model ID?

gemini-4-argon. It is a stable model on the Gemini API, available on the paid tier since October 13, 2026.

How much does the Gemini 4 Argon API cost?

$2 per million input tokens and $10 per million output tokens until November 30, 2026. From December 1 it is $4 and $20. Cached input is $0.10 per million now and $0.20 later, and Batch or Flex requests cost half.

When did the Gemini 4 Argon API launch?

October 13, 2026, two weeks after Google announced the model on September 30.

Is there a free tier for the Gemini 4 Argon API?

No. Google's pricing page says Gemini 4 Argon is not available on the free tier. You need a billing-enabled API project.

What is Gemini 4 Argon's context window?

1,048,576 tokens, shared between input and output. Output is capped at 65,536 tokens by default and can go up to the full window with long decoding.

Does Gemini 4 Argon work with the OpenAI SDK?

The Gemini API offers an OpenAI-compatible endpoint at generativelanguage.googleapis.com/v1beta/openai/, which you can point at gemini-4-argon. Test your exact use case before relying on it.

Should I wait for Gemini 4 Carbon instead?

No need. Carbon costs the same and is limited to cyber defenders for now. Build on Argon today and switch the model name when Carbon opens. Our sister site geminicarbon4.com tracks the Carbon API.

Sources: Google's Argon availability post, Gemini API models, Gemini API pricing, OpenAI compatibility, Gemini Enterprise Agent Platform release notes.