Verdict
GPT-6 Astra leads 7–3 across 12 shared benchmarks.
Gemini 4 Argon 3 · GPT-6 Astra 7 · 2 tied · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Unknown: GPT-6 Astra leads 2–0 across 3 shared unknown benchmarks (largest gap: Mystery Game Puzzles, 82.37 vs 60.34 — protocols undisclosed).
Coding: GPT-6 Astra leads 2–0 across 2 shared coding benchmarks (largest gap: FrontierSWE, 65.51 vs 54.96 — single evaluator).
Math: GPT-6 Astra leads 1–0 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 97.6 vs 78.05 — protocols undisclosed).
Reasoning: Gemini 4 Argon leads 2–0 across 2 shared reasoning benchmarks (largest gap: GPQA diamond, 95.2 vs 94.36 — protocols undisclosed).
GPT-6 Astra leads 7–3 across 12 shared benchmarks. Gemini 4 Argon leads 3 benchmarks and GPT-6 Astra leads 7, with 2 tied. Higher isn't always better — see the integrity caveats on each benchmark.
Gemini 4 Argon and GPT-6 Astra have 12 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-6 Astra leads unknown 2–0. GPT-6 Astra leads coding 2–0.