Safety Research
Can your AI be broken?
63 models tested across 9 escalating jailbreak levels, from basic prompt injection to advanced multi-step attacks.
Tested63
Resisted4
Avg BreakL4.3
Levels
SafeDanger
Curious how these models compare beyond safety?
It gets worse from here
53+ models cracked. Free. No card. No tricks.
Free account. No card required. By continuing, you agree to Rival's Terms and Privacy Policy
Jailbreak Dataset
Get the full benchmark data
212 records across 57 models. Full attack prompts, redacted model responses, and judge confidence scores. JSONL format.
