Model Leaderboard

Below is the leaderboard of all models on the site, ordered by ELO.

Main Stats Resilience by Request Type
Rank Model Company Country ELO Resilience # Tests # Jailbreaks Violent Crimes Non-Violent Crimes Sex Crimes Child Exploitation Defamation Specialized Advice Privacy Intellectual Property Indiscriminate Weapons Hate Self-Harm Sexual Content Elections Code Interpreter Abuse
πŸ₯‡ gpt-oss-20b OpenAI πŸ‡ΊπŸ‡Έ 911 98% 1098 21 97% 100% 95% 97% 100% 100% 99% 98% 100% 97% 95% 100% 95% 100%
πŸ₯ˆ gpt-oss-120b OpenAI πŸ‡ΊπŸ‡Έ 896 99% 1029 13 99% 99% 100% 98% 100% 100% 99% 99% 99% 99% 96% 99% 100% 97%
πŸ₯‰ qwen3-235b-a22b-instruct-2507 Alibaba πŸ‡¨πŸ‡³ 877 93% 1056 74 85% 92% 96% 97% 96% 99% 94% 96% 93% 94% 95% 95% 80% 94%
4 qwen3-32b Alibaba πŸ‡¨πŸ‡³ 821 84% 1062 165 82% 80% 85% 90% 85% 92% 79% 78% 93% 93% 92% 87% 68% 78%
5 kimi-k2-instruct-0905 Moonshot AI πŸ‡¨πŸ‡³ 780 79% 1167 241 68% 83% 72% 83% 84% 88% 70% 74% 89% 89% 84% 83% 69% 74%
6 mistral-small-3.2-24b-instruct-2506 Mistral πŸ‡«πŸ‡· 664 63% 984 368 51% 71% 61% 72% 57% 71% 52% 62% 79% 78% 74% 50% 40% 63%
7 mistral-nemo-instruct-2407 Mistral / Nvidia πŸ‡«πŸ‡· 650 62% 1053 404 57% 56% 60% 77% 64% 57% 64% 57% 57% 78% 73% 42% 65% 50%

Note: The statistics on this page are all as judged by Qwen3-32B. Qwen3-32B is not a perfect judge, meaning this represent a close approximation of LLM jailbreak resilience, rather than a perfect representation.