Google launches Gemini 4 Argon, a new flagship AI model aimed at improving physical reasoning for robotics and related development workflows. Google says Argon is designed to understand physical environments, generate code, and perform visual-semantic reasoning so developers can build and test robotic applications more directly from real-world inputs and simulations. The model is positioned as part of the Gemini 4 family and is rolled out for developer access through Google’s platforms.
Across outlets, reporting agrees that Argon becomes available globally for paying developers and businesses, with access through the Gemini API and Google AI Studio. Some coverage highlights that it supports Google Antigravity and other Google-integrated interfaces such as the Gemini app and AI Mode in Search, depending on the product tier. Pricing details in the coverage include an introductory rate of $2 per million input tokens and $10 per million output tokens, with cached input tokens offered at a steep discount. One outlet also reports a unified token limit of 1 million input/output tokens, reflecting how the model is configured for use.
While commentary outlets frame the launch as a strategic move to strengthen Google’s position in the AI market, the core factual details focus on the model’s robotics orientation, global developer availability, and the initial pricing and access paths Google provides.
