Google DeepMind presented Gemini 4 Carbon as its most advanced model to date, with a focus on cybersecurity and long-horizon tasks requiring numerous steps. The company cited results from the Artificial Analysis Intelligence Index and a separate evaluation focused on patching vulnerabilities. Internal testing reportedly began in early October 2026 on Google’s Jetski platform, comparing its coding performance with Anthropic’s Claude Opus 5.5. Early feedback praised its 3D scene generation but also highlighted slow output. Pricing, distribution channels and a clear launch timeline have not been disclosed.
Source: Read the original article


