Google tests new Gemini 4 AI models including Argon and Carbon

Google is currently conducting internal testing for its next generation of frontier AI models, specifically focusing on the Gemini 4 family. The company is preparing to roll out the Gemini 4 Argon model to the public, which is being positioned as a successor to previous Gemini iterations.

Internal documents and employee reports indicate that Google is also testing even more advanced versions, such as a model codenamed Carbon, via its internal coding platform, Jetski. While early versions of Argon received positive feedback, some employees noted they initially lagged behind competitors like Anthropic in complex coding tasks. The testing of newer iterations like Carbon suggests an effort to close the gap with rivals such as OpenAI and Anthropic in the race for highly capable AI coding agents.