CWE-bench v1

Security vulnerability remediation; pass@1.

The best 1 · 1 models tested

Top 1 · best setting per model · 1 models, 1 results (0 independent)

Measured by independent testersIndependent Reported by the makerVendor-reported

CWE-bench v1 leaderboard
				Thinking levelSetting
1	Gemini 4 Argon	Google (Gemini / DeepMind)	68.0%	Maxhighest thinking settings	Maker's own figureVendor-reportedGoogle ↗	—	30 Sep 2026