benchgap
Meta · model

Muse Spark 1.3 (max) benchmark scores

As of 2026-10-07, Muse Spark 1.3 (max) (Meta) has measured scores on 11 benchmarks and estimated scores on 4 more.

BenchmarkScoreSource
Terminal-Bench 2.184.3%measured
Terminal-Bench 4.033.3%measured
Terminal-Bench Science 0.111.0%measured
GPQA Diamond93.5%measured
SciCode58.8%measured
CritPt24.9%measured
GDP.pdf26.6%measured
τ³-Bench Banking50.5%measured
MMMU-Pro85.0%estimated ± 2.6 pp, high confidence
AA-AnalystAgent45.6%estimated ± 7.8 pp, medium confidence
Harvey LAB89.3%estimated ± 1.6 pp, low confidence
Humanity's Last Exam48.7%measured
AutomationBench57.9%measured
EnterpriseOps-Gym42.8%estimated ± 5.3 pp, medium confidence
ITBench SRE33.2%measured