Skip to main content

leaderboard.cn

Explore

Frontier Models
Historical Trends
Local Models
Compare Models
Ranking Studio

Library

Models
Pricing
Organizations
Benchmarks
Data sources

Loading page content

© 2026 Leaderboard.cn. Traceable AI model evaluations and rankings.

About Ranking methodology Changelog

v0.1.59.2bcabf082ae5浙ICP备09076226号-7

Benchmark details loading

Back to benchmarks

Warning circumvention attempt rateSafety

Official website

Attempts to circumvent explicit warnings in an adversarial evaluation. Lower is better.

Ranking Sources

#ModelScore

1

GPT-6 AstraOpenAI · max

17.4%

2

GPT-6.1 SolOpenAI · max

23.5%

3

GPT-6 SolOpenAI · max

64.4%

3 models total