Hugging Face Blog·· 1 d ago
CWE-bench v2: More coverage, quality, and transparency
CWE-bench v2: More coverage, quality, and transparency
AI summary
CWE-bench is designed to test the limits of frontier cyber capabilities. In our continued pursuit to push the horizon at Collinear AI, CWE-bench v2 opens up a public set of 100 tasks that anyone can run and evaluate their models on.
Selection record
Threshold 60Official, first-handFirst 42Second 52
Not admittedSum of both 94 < twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：AI智能体安全基准评测与模型排名
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Hugging Face Blog · huggingface.co