Skip to main content

Gemini 4 Argon: 1M Token Output, Built for the Long Haul

Google DeepMind's Gemini 4 Argon: A Model That Doesn't Quit

What if an AI could write a million tokens in one shot? That's exactly what Google DeepMind's new Gemini 4 Argon does. Launched on September 30, it blows past the typical 128K output limit of most cutting-edge models, allowing it to generate entire codebases, lengthy technical reports, or multi-step automation scripts without breaking them into chunks.

Image

Not Your Average Chatbot

Argon isn't designed for casual chit-chat. It's built for what Google calls "complex, long-term workflows"—think software engineering, law, finance, and cybersecurity defense. The model is being rolled out in phases: first to trusted cybersecurity professionals, then to developers, enterprises, and subscription users.

The benchmarks are impressive. On DeepSWE v1.1, which tests long-horizon software engineering tasks, Argon hit a record 77.9%. It also topped the Vals Index (a weighted mix of finance, programming, law, and taxation) and scored 51.3% on Zapier's AutomationBench. For long video understanding, it reached 91.7% on LVBench—the highest yet. And visually, it can handle professional charts, extract details from lengthy videos, and perform cross-document actions.

Security That Puts Up a Fight

Cybersecurity is at the heart of this release. Argon is trained to autonomously discover, verify, and fix critical vulnerabilities. For trusted defenders and Google's internal teams, there's even a version with "no network barrier." In one case, security vendor Wiz used Argon through the Scan for Good project to uncover a serious flaw in a global hospital's medical software—something previous cutting-edge models had missed. On CWE-bench v1, which tests vulnerability repair, Argon ranked first with a 68% score.

Pricing and Defenses

During the introductory period, Argon costs $2 per million input tokens and $10 per million output tokens; later, those prices will double to $4 and $20. Before wide release, Google is beefing up four advanced defenses: preventing abuse, blocking prompt injection, monitoring misaligned reasoning chains, and strengthening sandboxes. Thousands of Google employees are already using Argon to debug code, design algorithms, and migrate large codebases. In one case, quantum algorithms were improved by 40% in minutes.

In short, Argon isn't here to chat. It's here to tackle the hardest challenges professionals face.

Key Points:

  • Gemini 4 Argon outputs up to 1 million tokens in a single go, far beyond the typical 128K limit.
  • Targets complex, long-term workflows in software engineering, law, finance, and cybersecurity.
  • Tops benchmarks: 77.9% on DeepSWE v1.1, 51.3% on AutomationBench, 91.7% on LVBench.
  • Cybersecurity focus: autonomously finds and fixes vulnerabilities; already used to spot a hospital software flaw.
  • Pricing: $2/$10 per million input/output tokens initially, rising to $4/$20 later.