benchgap
Zhipu (GLM) · model

GLM-4.5 benchmark scores

As of 2026-11-22, GLM-4.5 (Zhipu (GLM)) has measured scores on 4 benchmarks and estimated scores on 24 more.

BenchmarkScoreSource
AA-GPQA Diamond74.4%estimated ± 3.6 pp, high confidence
AA-HLE9.3%estimated ± 6.2 pp, medium confidence
AA-SciCode43.8%estimated ± 5.1 pp, low confidence
Artificial Analysis Intelligence Index12.6%estimated ± 5.8 pp, medium confidence
CritPt0.9%estimated ± 5.3 pp, medium confidence
GPQA Diamond80.1%estimated ± 1.8 pp, medium confidence
HLE29.7%estimated ± 9.2 pp, low confidence
LiveCodeBench v680.4%estimated ± 2.2 pp, low confidence
AA-Omniscience Accuracy25.9%estimated ± 6.5 pp, medium confidence
AA Coding Index45.7%estimated ± 11.5 pp, low confidence
SWE-bench Verified57.2%estimated ± 13.1 pp, low confidence
SWE-bench Pro57.2%estimated ± 4.7 pp, medium confidence
MMLU-Pro82.6%estimated ± 4.8 pp, high confidence
GPQA75.1%estimated ± 4.5 pp, high confidence
MMLU90.0%estimated ± 1.2 pp, high confidence
Vals GPQA Diamond72.2%measured
Vals LiveCodeBench67.4%measured
Vals MMLU-Pro81.2%measured
AA MMLU-Pro84.8%estimated ± 1.1 pp, high confidence
ARC-AGI-149.3%estimated ± 6.5 pp, low confidence
ARC-AGI-20.0%estimated ± 13.8 pp, low confidence
CursorBench 3.259.1%estimated ± 7.2 pp, low confidence
Mystery Game Puzzles4.6%estimated ± 4.7 pp, medium confidence
VulcanBench v377.5%estimated ± 5.6 pp, low confidence
HLE w/o tools25.2%estimated ± 4.9 pp, high confidence
ARC-AGI-30.0%estimated ± 8.6 pp, low confidence
CompileBench48.9%measured
NL2Repo45.7%estimated ± 7.9 pp, low confidence