Verdict
Claude Opus 5 leads 10–4 across 14 shared benchmarks.
Claude Opus 5 10 · GPT-5.6 Terra 4 · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Reasoning: Claude Opus 5 leads 3–0 across 3 shared reasoning benchmarks (largest gap: SimpleQA Verified, 56.7 vs 43.1 — single evaluator).
Coding: Claude Opus 5 leads 2–0 across 2 shared coding benchmarks (largest gap: FrontierCode, 53.4 vs 41.3 — single evaluator).
Math: Evenly split 1–1 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 73.17 vs 70.73 — single evaluator).
Unknown: Claude Opus 5 leads 2–0 across 2 shared unknown benchmarks (largest gap: ProofBench, 99 vs 71 — single evaluator).
Claude Opus 5 leads 10–4 across 14 shared benchmarks. Claude Opus 5 leads 10 benchmarks and GPT-5.6 Terra leads 4. Higher isn't always better — see the integrity caveats on each benchmark.
Claude Opus 5 and GPT-5.6 Terra have 14 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
Claude Opus 5 leads reasoning 3–0. Claude Opus 5 leads coding 2–0.