Verdict
GPT-6 Astra leads 15–0 across 16 shared benchmarks.
GPT-5.6 Sol 0 · GPT-6 Astra 15 · 1 tied · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Coding: GPT-6 Astra leads 3–0 across 3 shared coding benchmarks (largest gap: MirrorCode, 46.7 vs 20 — protocols undisclosed).
Reasoning: GPT-6 Astra leads 3–0 across 3 shared reasoning benchmarks (largest gap: SimpleQA Verified, 75.6 vs 69.7 — protocols undisclosed).
Unknown: GPT-6 Astra leads 3–0 across 3 shared unknown benchmarks (largest gap: EBR-bench, 76.19 vs 44.76 — protocols undisclosed).
Math: GPT-6 Astra leads 1–0 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 97.6 vs 82.93 — protocols undisclosed).
GPT-6 Astra leads 15–0 across 16 shared benchmarks. GPT-5.6 Sol leads 0 benchmarks and GPT-6 Astra leads 15, with 1 tied. Higher isn't always better — see the integrity caveats on each benchmark.
GPT-5.6 Sol and GPT-6 Astra have 16 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-6 Astra leads coding 3–0. GPT-6 Astra leads reasoning 3–0.