Recent leaks from early November 2026 testers describe an internal Google DeepMind checkpoint called Gemini Krypton as delivering roughly twice the code output and stronger reasoning than Gemini 4 Carbon on complex prompts. Carbon itself debuted near the top of the Text Arena leaderboard in late October with scores around 1526-1538, ahead of rivals like Claude Opus 5.5. These unverified reports have driven trader consensus toward a strong Krypton debut, pricing top-10 odds above 90 percent while leaving top-1 closer to even amid the fast Gemini 4 release cadence. Key upcoming catalysts include any official announcement, API listing, or Arena appearance, though timelines remain uncertain and product shifts are common in frontier large language model development.