benchgap
Zhipu (GLM) · model

GLM-5.3-Flash benchmark scores

As of 2026-10-07, GLM-5.3-Flash (Zhipu (GLM)) has measured scores on 12 benchmarks and estimated scores on 3 more.

BenchmarkScoreSource
Terminal-Bench 2.184.3%measured
Terminal-Bench 4.032.8%measured
Terminal-Bench Science 0.14.8%measured
GPQA Diamond91.2%measured
SciCode51.6%measured
CritPt15.4%measured
GDP.pdf15.4%measured
τ³-Bench Banking47.2%measured
MMMU-Pro76.6%estimated ± 2.6 pp, high confidence
AA-AnalystAgent13.5%estimated ± 3.6 pp, medium confidence
Harvey LAB88.8%estimated ± 1.2 pp, medium confidence
Humanity's Last Exam39.9%measured
AutomationBench60.4%measured
EnterpriseOps-Gym33.2%measured
ITBench SRE51.2%measured