So who actually won? Final score across 10 brutal tasks: Gemini 3.0 Pro: 6 wins
Claude 4.5 Sonnet: 3 wins
Ties: 1 (both were solid) Gemini crushed the hard reasoning, system debugging, and multi-step planning. Claude shined on storytelling, cultural nuance, and softer human
Comparative Performance Analysis of Gemini 3.0 Pro and Claude 4.5 Sonnet
By
–