What does Google's new Gemini 4 Argon model actually do for developers and researchers?
Google has released Gemini 4 Argon, its latest frontier AI model, to developers, enterprises, and subscribers worldwide. The model carries a 1 million token context window and an introductory price of $2 per million input tokens and $10 per million output tokens.
The model and its price
Gemini 4 Argon is a large language model that can process text, code, and images. It launched globally on October 13, 2026. Developers and enterprises access it through the API in AI Studio and the Gemini platform. Google AI Ultra subscribers can use it in Google Antigravity. Both Ultra and Pro subscribers get it in the Gemini app and AI Mode in Search. The introductory pricing runs until November 30, 2026. After that the rate doubles to $4 per million input tokens and $20 per million output tokens. Cached input tokens cost 95 percent less during the introductory period. The model supports a unified input-output limit of 1 million tokens. It works.
Engineering work inside Google
Google's own teams have been using Argon on production problems for weeks. Software engineers shipping code through Antigravity saw agentic code submissions rise by more than 35 percent. Code rollbacks fell by nearly a third. The Chrome team used Argon to untangle a ~6 million line legacy folder inside Chromium's 40 million line codebase. Argon piped multiple clean changes at once and finished an architectural refactor across hundreds of files in weeks instead of months.
The sources this article was written from
- "Google AI blog
https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-4-argon/"