# Gemini 4 Argon

Google's flagship reasoning model for complex software engineering, deep mathematical and scientific problem solving, and autonomous multi-step agents. Multimodal input accepting text, images, video, audio, and documents/PDFs.

## Overview

- **Provider:** Google
- **Model ID:** `gemini-4-argon`
- **Type:** Chat
- **Canonical URL:** https://modelstack.cc/models/chat/gemini-4-argon
- **Version:** 4 Argon
- **Best for:** Complex software engineering, math and science reasoning, autonomous multi-step agents, multimodal analysis

## Pricing

| Usage | Price |
| --- | ---: |
| Input | $4 / 1M tokens |
| Output | $20 / 1M tokens |
| Cache read | $0.2 / 1M tokens |
| Cache write (5 minutes) | $0.375 / 1M tokens |

## Specifications

- **Context window:** 1M
- **Maximum output:** 943K
- **Released:** 2026-10-13

## Capabilities

- chat
- tool_use
- reasoning
- coding
- analysis

## API usage

Use this model ID in a ModelStack API request:

```json
{
  "model": "gemini-4-argon"
}
```

For the rendered page, visit https://modelstack.cc/models/chat/gemini-4-argon.