Verdict
Evenly split 4–4 across 10 shared benchmarks.
GPT-4 (Jun 2023) 4 · GPT-4o mini 4 · 2 tied · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Math: GPT-4o mini leads 3–0 across 3 shared math benchmarks (largest gap: MATH level 5, 52.63 vs 22.97 — single evaluator).
Unknown: GPT-4 (Jun 2023) leads 2–0 across 3 shared unknown benchmarks (largest gap: DTBench, 37.78 vs 24 — single evaluator).
Evenly split 4–4 across 10 shared benchmarks. GPT-4 (Jun 2023) leads 4 benchmarks and GPT-4o mini leads 4, with 2 tied. Higher isn't always better — see the integrity caveats on each benchmark.
GPT-4 (Jun 2023) and GPT-4o mini have 10 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-4o mini leads math 3–0. GPT-4 (Jun 2023) leads unknown 2–0.