Skip to main content

Google's Argon AI Opens Up, But Can It Code?

Google's Argon AI Opens Up, But Can It Code?

Google announced Tuesday that its flagship AI model, Argon, is now available to developers and enterprise customers worldwide. Previously, it was offered only to a select group of trusted cybersecurity professionals through Google's Fairwind program. This phased rollout aims to balance security assessments with commercial promotion—a clear sign that Google is treading carefully when expanding access to cutting-edge models.

Image

Starting now, paying users can access Argon through the Google AI Studio API, Google AI Ultra and Pro subscription services in the Gemini App, and the intelligent agent development platform Antigravity. The pricing? $2 per million input tokens and $10 per million output tokens. But hold on—prices double starting in December.

Even after the hike, Argon remains cheaper than OpenAI's GPT-6Astra (mentioned in the text), which charges $10 per million input tokens and $50 per million output tokens. That's a big difference. However, according to reports, some Google employees still have doubts about Argon's performance in areas like coding. So the question lingers: is Argon really as good as Google claims?

Koray Kavukcuoglu, head of Google DeepMind, is optimistic. He says internal software engineers have used Argon to deliver high-quality code more quickly, and the model has also been used to optimize the user experience of Chrome. Google hopes to demonstrate the model's value in real engineering tasks. Kavukcuoglu also notes that Argon combines multimodal reasoning, agent coding, and physical understanding capabilities—potentially advancing embodied robotics research.

In one demo, researchers reconstructed the scene layout and motion trajectory from videos of humans throwing sandbags. They then passed the computational results to a humanoid robot, which hit the target seven times consecutively in a real environment. That's impressive—but does it translate to everyday coding tasks? The jury is still out.

For developers, the opportunity is clear: Argon is now within reach. But with pricing set to double and skepticism from within Google's own ranks, it's worth watching how the model performs under real-world pressure. After all, a robot hitting a target seven times is cool, but can it debug your code at 2 a.m.? That's the real test.

Key Points

  • Global availability: Argon is now open to developers and enterprises worldwide via Google AI Studio API, Gemini App subscriptions, and Antigravity.
  • Pricing: $2 per million input tokens and $10 per million output tokens, doubling in December.
  • Competitive edge: Still cheaper than OpenAI's GPT-6Astra, but internal doubts about coding performance persist.
  • Real-world demos: Argon showed promise in robotics, but practical coding value remains to be proven.