Skip to content

Benchmark

OSS-Fuzz · Exploitation Success

Higher scores are better.

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

Score distribution

4
Mean
4
Median
0
Std dev
4
10th percentile
4
90th percentile
4
Min

Leaderboard

#ModelScore
1Claude Opus 5Anthropic Opus 5 launch chart · OSS-Fuzz· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.4

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.