benchgap
Alibaba (Qwen) · model

Qwen2.5 Coder 32B Instruct benchmark scores

As of 2026-10-07, Qwen2.5 Coder 32B Instruct (Alibaba (Qwen)) has measured scores on 3 benchmarks and estimated scores on 26 more.

BenchmarkScoreSource
AA-GPQA Diamond41.7%measured
AA-HLE3.5%measured
Artificial Analysis Intelligence Index6.7%measured
CritPt0.3%estimated ± 2.3 pp, high confidence
GPQA Diamond68.2%estimated ± 2.2 pp, medium confidence
HLE19.4%estimated ± 9.9 pp, low confidence
AA-Omniscience Accuracy17.3%estimated ± 6.9 pp, medium confidence
BioMysteryBench (human-difficult)35.2%estimated ± 8.1 pp, low confidence
MMLU-Pro66.5%estimated ± 3.9 pp, medium confidence
GPQA42.1%estimated ± 3.2 pp, medium confidence
ARC-AGI-10.3%estimated ± 5.4 pp, low confidence
ARC-AGI-20.2%estimated ± 14.2 pp, low confidence
Vals GPQA Diamond28.6%estimated ± 7.4 pp, low confidence
Vals MMLU-Pro83.1%estimated ± 2.5 pp, medium confidence
HLE w/o tools2.7%estimated ± 5.7 pp, low confidence
C-Eval77.1%estimated ± 1.0 pp, low confidence
MMLU-Redux71.0%estimated ± 2.1 pp, low confidence
SuperGPQA33.3%estimated ± 10.4 pp, low confidence
HealthBench Hard0.0%estimated ± 7.7 pp, low confidence
MMLU-Pro (Arcee)77.2%estimated ± 5.4 pp, low confidence
ARC-AGI-31.6%estimated ± 10.7 pp, low confidence
BioMysteryBench (human-solvable)0.1%estimated ± 1.0 pp, low confidence
HealthBench (raw)5.5%estimated ± 6.4 pp, low confidence
HealthBench (length-adjusted)53.9%estimated ± 3.7 pp, low confidence
HealthBench Professional57.7%estimated ± 4.8 pp, low confidence
HealthBench Professional (raw)20.0%estimated ± 5.0 pp, low confidence
HLE-Verified15.3%estimated ± 2.0 pp, low confidence
LABBench224.8%estimated ± 1.8 pp, low confidence
MMLU77.9%estimated ± 1.9 pp, low confidence