benchgap
Other · model

ZAYA1-8B benchmark scores

As of 2026-11-22, ZAYA1-8B (Other) has measured scores on 10 benchmarks and estimated scores on 34 more.

BenchmarkScoreSource
AA-GPQA Diamond68.6%estimated ± 4.4 pp, high confidence
AA-HLE10.1%estimated ± 3.9 pp, high confidence
AA-SciCode27.0%estimated ± 4.7 pp, high confidence
AIME2689.1%measured
Artificial Analysis Intelligence Index10.1%estimated ± 7.4 pp, medium confidence
CritPt0.3%estimated ± 4.0 pp, high confidence
GDPval-AA0.0%estimated ± 5.5 pp, low confidence
GPQA Diamond71.0%measured
HLE9.6%estimated ± 6.5 pp, medium confidence
IFBench52.6%measured
LiveCodeBench v665.8%measured
AA-Omniscience Accuracy10.6%estimated ± 8.4 pp, medium confidence
SciCode26.5%estimated ± 4.9 pp, high confidence
IFEval85.6%measured
τ²-bench results63.2%estimated ± 12.4 pp, low confidence
AA Coding Index18.3%estimated ± 5.7 pp, medium confidence
SWE-bench Verified57.5%estimated ± 9.7 pp, low confidence
BFCL v439.2%measured
SWE-bench Pro9.4%estimated ± 6.0 pp, low confidence
MMLU-Pro74.2%measured
AA-IFBench56.7%estimated ± 7.4 pp, medium confidence
BBH92.4%estimated ± 2.8 pp, medium confidence
GPQA71.0%measured
MMLU86.1%estimated ± 2.1 pp, high confidence
Vals GPQA Diamond68.9%estimated ± 4.5 pp, high confidence
Vals MMLU-Pro79.4%estimated ± 2.3 pp, high confidence
MATH-50096.8%estimated ± 3.1 pp, medium confidence
MMMLU75.7%estimated ± 1.1 pp, low confidence
AIME 202589.9%estimated ± 1.8 pp, medium confidence
AA MMLU-Pro85.3%estimated ± 1.2 pp, low confidence
ARC-AGI-123.1%estimated ± 6.7 pp, low confidence
ARC-AGI-20.0%estimated ± 10.9 pp, low confidence
FrontierMath v2 (Tiers 1-3)5.9%estimated ± 8.8 pp, low confidence
HLE w/o tools9.4%estimated ± 4.3 pp, medium confidence
C-Eval88.8%estimated ± 1.2 pp, high confidence
HMMT Feb 202671.6%measured
HMMT Nov 202580.1%estimated ± 2.8 pp, medium confidence
MMAnswerBench73.2%estimated ± 1.3 pp, medium confidence
MMLU-Redux86.9%estimated ± 0.6 pp, high confidence
SuperGPQA50.4%estimated ± 8.5 pp, low confidence
MMLU-Pro (Arcee)79.4%estimated ± 3.4 pp, medium confidence
SimpleQA23.7%estimated ± 11.7 pp, low confidence
IMOAnswerBench59.3%measured
PinchBench43.5%estimated ± 5.0 pp, low confidence