Verdict
Qwen 3.8 Max leads 9–1 across 10 shared benchmarks.
GPT-5 1 · Qwen 3.8 Max 9 · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Unknown: Qwen 3.8 Max leads 4–0 across 4 shared unknown benchmarks (largest gap: ProofBench, 58 vs 18 — single evaluator).
Math: Qwen 3.8 Max leads 2–0 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 46.34 vs 21.95 — single evaluator).
Reasoning: Evenly split 1–1 across 2 shared reasoning benchmarks (largest gap: GPQA diamond, 90.24 vs 81.57 — single evaluator).
Qwen 3.8 Max leads 9–1 across 10 shared benchmarks. GPT-5 leads 1 benchmark and Qwen 3.8 Max leads 9. Higher isn't always better — see the integrity caveats on each benchmark.
GPT-5 and Qwen 3.8 Max have 10 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
Qwen 3.8 Max leads unknown 4–0. Qwen 3.8 Max leads math 2–0.