Gemini 4 Argon Is Here, But Free Users Get Left Behind

Google has finally rolled out its flagship frontier model, Gemini 4 Argon, to the public. According to announcements detailed by The Verge and Ars Technica, the model targets complex real-world workflows in coding, law, and science, but it comes with a strict paywall.

Who actually gets access to Argon?

Free users and those on basic plus plans are completely locked out of the new architecture. Instead, free tiers are restricted strictly to lighter flash models, as reported by The Decoder.

To use Argon in the consumer Gemini app or Search's AI mode, you need at least the AI Pro plan at $20/month. If you want to run Argon inside Google Antigravity for multi-threaded agentic coding, you must upgrade to the AI Ultra plan, which starts at $99/month, according to Ars Technica's breakdown.

How to test or integrate the model

Developers looking to build on top of Gemini 4 Argon can access it directly via the API in AI Studio and the Gemini platform. Google is running an introductory pricing tier through November 30.

• API Pricing (Promo): $2 per million input tokens and $10 per million output tokens.
• Cached Inputs: Enjoy a 95% discount on cached input tokens during the promotional window.
• Post-Promo Rates: Prices will double on December 1 to $4 per million input tokens and $20 per million output tokens.

Vale a pena para a sua rotina?

If your daily work involves heavy codebase refactoring, enterprise data synthesis, or complex multi-step reasoning, Argon delivers tangible efficiency. Google notes internal tests where the model cut session preparation time by 40% and accelerated open-source Gemma inference speeds by 2.85x on an A100 GPU. However, for casual users or light prompt writers, the jump to a $20 or $99 monthly subscription makes it a tough sell compared to existing alternatives.