Gemini 4 Carbon vs GPT-6 Astra
- Score
- Gemini 4 Carbon 6 · GPT-6 Astra 7 · 2 tied — higher isn't always better; check the caveats.
- Shared
- 15 benchmarks measured for both models
- Integrity
- 13 of 15 graded A or B
Reported scores — protocols may differ. How we compare
Each bar is drawn on its benchmark's own scale — the ceiling the record states where it has one (of 100), otherwise the best score recorded on that benchmark (rel. 2103). The margin is in points: + is Gemini 4 Carbon, - is GPT-6 Astra.
Game Playing: Gemini 4 Carbon leads 2–1 across 3 shared game playing benchmarks (largest gap: Mystery Game Puzzles, 82.37 vs 70.26 — protocols undisclosed).
Reasoning: Gemini 4 Carbon leads 3–0 across 3 shared reasoning benchmarks (largest gap: SimpleBench, 91.12 vs 83.8 — single evaluator).
Abstract Reasoning: GPT-6 Astra leads 1–0 across 2 shared abstract reasoning benchmarks (largest gap: ARC-AGI-2, 95 vs 94.17 — protocols undisclosed).
Math: GPT-6 Astra leads 1–0 across 2 shared math benchmarks (largest gap: FrontierMath-Tier-4-v2-Private, 97.6 vs 95.12 — protocols undisclosed).
GPT-6 Astra leads 7–6 across 15 shared benchmarks. Gemini 4 Carbon leads 6 benchmarks and GPT-6 Astra leads 7, with 2 tied. Higher isn't always better — see the integrity caveats on each benchmark.
Gemini 4 Carbon and GPT-6 Astra have 15 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
Gemini 4 Carbon leads reasoning 3–0. Gemini 4 Carbon leads game playing 2–1.