Skip to main content
leaderboard
.cn
Toggle Sidebar
Search models or organizations…
⌘K
Explore
Frontier Models
Historical Trends
Local Models
Compare Models
Ranking Studio
Library
Models
Pricing
Organizations
Benchmarks
Data sources
Settings
Toggle Sidebar
leaderboard
.cn
Loading page content
Feedback
Benchmark details loading
Back to benchmarks
CWE-bench v1
Coding
Agents
Share
Official website
Version 1 of a security vulnerability remediation benchmark. Scores report pass@1; the official leaderboard breaks ties using pass@4.
Ranking
Sources
#
Model
Score
1
Gemini 4 Argon
Google
68.0
%
2
GPT-6 Astra
OpenAI
68.0
%
3
Claude Opus 5.5
Anthropic
67.0
%
4
Claude Fable 5.1
Anthropic
58.0
%
4 models total
