ExploitBench

Develop working exploits for real vulnerabilities; average capability-coverage score.

The best 2 · 2 models tested

Top 2 · best setting per model · 2 models, 2 results (0 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported

ExploitBench leaderboard
				Thinking levelSetting
1	GLM-5.3	Z.ai (Zhipu)	54.4%	Maxreasoning_effort=max	Maker's own figureVendor-reportedZ.ai ↗	Claude Code 2.1.207	14 Aug 2026
2	GLM-5.2	Z.ai (Zhipu)	24.4%	Maxsetting not stated (GLM-5.3 blog comparison column)	Maker's own figureVendor-reportedZ.ai ↗	—	14 Aug 2026