Verdict
GPT-4 (Mar 2023) leads 4–0 across 4 shared benchmarks.
GPT-4 (Mar 2023) 4 · Qwen2.5-Coder-32B 0 · higher isn't always better — check the caveats below.
Reported scores — protocols may differ. How we compare
Commonsense Reasoning: GPT-4 (Mar 2023) leads 2–0 across 2 shared commonsense reasoning benchmarks (largest gap: HellaSwag, 93.73 vs 77.33 — single evaluator).
GPT-4 (Mar 2023) leads 4–0 across 4 shared benchmarks. GPT-4 (Mar 2023) leads 4 benchmarks and Qwen2.5-Coder-32B leads 0. Higher isn't always better — see the integrity caveats on each benchmark.
GPT-4 (Mar 2023) and Qwen2.5-Coder-32B have 4 benchmarks in common in our data — those are the only rows where a direct, apples-to-apples comparison is drawn.
GPT-4 (Mar 2023) leads commonsense reasoning 2–0.