Side-by-side benchmark comparison across knowledge, coding, math, and reasoning.
Gemini 3 Pro Deep Think wins overall with a score of 81 vs 77 (4 point difference).Gemini 3 Pro Deep Think wins 2 out of 4 categories.
Gemini 3 Pro Deep Think
96
GPT-5.1-Codex-Max
95
Gemini 3 Pro Deep Think
91
GPT-5.1-Codex-Max
94
Gemini 3 Pro Deep Think
97.1
GPT-5.1-Codex-Max
97.1
Gemini 3 Pro Deep Think
94
GPT-5.1-Codex-Max
93
Gemini 3 Pro Deep Think scores higher overall with 81 vs 77, a difference of 4 points across all benchmarks.
Gemini 3 Pro Deep Think leads in knowledge tasks with an average score of 96 vs 95.
GPT-5.1-Codex-Max leads in coding with an average score of 94 vs 91.
Gemini 3 Pro Deep Think and GPT-5.1-Codex-Max are tied for math with average scores of 97.1.
Gemini 3 Pro Deep Think leads in reasoning with an average score of 94 vs 93.