Verdict
GPT-5 leads 21–2 across 23 shared benchmarks.
Gemini 2.5 Pro (Jun 2025) 2 · GPT-5 21 · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Reasoning: GPT-5 leads 4–2 across 6 shared reasoning benchmarks (largest gap: ARC-AGI, 65.7 vs 41 — single evaluator).
Agentic: GPT-5 leads 2–0 across 2 shared agentic benchmarks (largest gap: APEX-Agents, 18.3 vs 6.6 — protocols undisclosed).
Coding: GPT-5 leads 2–0 across 2 shared coding benchmarks (largest gap: Terminal Bench, 49.6 vs 32.6 — single evaluator).
Math: GPT-5 leads 2–0 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 21.95 vs 0 — single evaluator).
GPT-5 leads 21–2 across 23 shared benchmarks. Gemini 2.5 Pro (Jun 2025) leads 2 benchmarks and GPT-5 leads 21. Higher isn't always better — see the integrity caveats on each benchmark.
Gemini 2.5 Pro (Jun 2025) and GPT-5 have 23 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-5 leads reasoning 4–2. GPT-5 leads agentic 2–0.