Skip to content

Benchmark

CWE-bench v1

Scores are percentages. Higher is better.

Models measured
1
Scores recorded
1
Sources
1
IQ weight
No weight published

Score distribution

68%
Mean
68%
Median
0%
Std dev
68%
10th percentile
68%
90th percentile
68%
Min

Leaderboard

#ModelScore
1Gemini 4 Argonv1 · vulnerability remediation · tied first· Reported by Google DeepMind· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.68%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.