benchgap
DeepSeek · model

DeepSeek R1 Distill Qwen 32B benchmark scores

As of 2026-11-22, DeepSeek R1 Distill Qwen 32B (DeepSeek) has measured scores on 5 benchmarks and estimated scores on 29 more.

BenchmarkScoreSource
AA-GPQA Diamond61.5%measured
AA-HLE4.6%measured
Artificial Analysis Intelligence Index8.4%measured
CritPt0.2%estimated ± 3.4 pp, high confidence
GPQA Diamond69.1%estimated ± 2.4 pp, medium confidence
HLE16.8%estimated ± 8.5 pp, medium confidence
IFBench53.6%estimated ± 6.6 pp, low confidence
AA-LCR8.7%measured
AA-Omniscience Accuracy19.0%estimated ± 7.3 pp, medium confidence
IFEval75.2%estimated ± 2.8 pp, high confidence
LongBench v251.2%estimated ± 5.5 pp, low confidence
BioMysteryBench (human-difficult)32.4%estimated ± 8.3 pp, low confidence
MMLU-Pro76.2%estimated ± 5.4 pp, medium confidence
AA-IFBench22.9%measured
GPQA64.7%estimated ± 6.1 pp, medium confidence
MMLU83.3%estimated ± 3.8 pp, high confidence
Vals GPQA Diamond60.6%estimated ± 3.2 pp, high confidence
Vals MMLU-Pro75.5%estimated ± 3.0 pp, high confidence
MMMLU82.1%estimated ± 1.7 pp, high confidence
AA MMLU-Pro83.4%estimated ± 1.2 pp, medium confidence
ARC-AGI-172.6%estimated ± 7.9 pp, low confidence
ARC-AGI-226.1%estimated ± 14.4 pp, low confidence
HLE w/o tools22.6%estimated ± 4.9 pp, high confidence
ARC-AGI-31.3%estimated ± 9.8 pp, low confidence
ARC-AGI-3 Provider Adapter2.7%estimated ± 7.3 pp, low confidence
C-Eval88.1%estimated ± 1.0 pp, low confidence
MMLU-Redux89.1%estimated ± 1.2 pp, low confidence
SuperGPQA61.7%estimated ± 10.1 pp, low confidence
HealthBench (length-adjusted)54.1%estimated ± 3.7 pp, medium confidence
HealthBench Professional56.7%estimated ± 4.2 pp, low confidence
SimpleQA19.3%estimated ± 7.8 pp, medium confidence
MRCRv222.8%estimated ± 11.3 pp, low confidence
MMLongBench-Doc42.2%estimated ± 2.6 pp, low confidence
BioBasicsBench10.7%estimated ± 10.2 pp, low confidence