The development
What happened
Google released Gemini 4 Argon on October 13 through developer APIs and several paid consumer services. The model is positioned for sustained coding, professional and scientific work, with introductory API prices and a combined input-output token budget. Google also reports internal engineering results from its use in Antigravity, giving an early account of practical deployment alongside its model-performance claims.
Key facts
Developers and enterprises can access Argon through APIs in AI Studio and the Gemini platform. Antigravity access is offered to Google AI Ultra subscribers.
The Gemini app and AI Mode offer Argon to Google AI Ultra and Pro subscribers; availability should not be read as unrestricted free access.
Introductory API pricing is $2 per million input tokens and $10 per million output tokens, with a 95% discount for cached input.
The Gemini API supports a one-million-token unified input-output limit. Input context and generated output share that budget.
Google says teams using Argon in Antigravity increased agentic code submissions by more than 35% and reduced rollbacks by nearly one-third. These are internal company results, not controlled external productivity estimates.
The context
Why it matters
TideReach assessment: The combination of broad access channels, long context and published token prices makes Argon a concrete alternative for extended AI workflows. Real costs depend on how much context is reused, how much output a task generates and how often work needs correction. The relevant comparison is useful task completion with acceptable review effort; internal coding figures alone cannot establish the same gains for other organizations or domains.