Verdict
GPT-5.2 leads 12–4 across 17 shared benchmarks.
Gemini 3 Flash 4 · GPT-5.2 12 · 1 tied · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Reasoning: Evenly split 2–2 across 4 shared reasoning benchmarks (largest gap: SimpleQA Verified, 67.4 vs 38.9 — single evaluator).
Math: GPT-5.2 leads 2–0 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 31.7 vs 17.07 — single evaluator).
Unknown: Gemini 3 Flash leads 1–0 across 2 shared unknown benchmarks (largest gap: Mystery Game Puzzles, 18.48 vs 15.18 — single evaluator).
GPT-5.2 leads 12–4 across 17 shared benchmarks. Gemini 3 Flash leads 4 benchmarks and GPT-5.2 leads 12, with 1 tied. Higher isn't always better — see the integrity caveats on each benchmark.
Gemini 3 Flash and GPT-5.2 have 17 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-5.2 leads math 2–0. Gemini 3 Flash leads unknown 1–0.