Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
JailbreakEvalpypi.org · Low evidenceJailbreakEval is a Python package for evaluating jailbreak attacks on large language models. It provides standardized metrics and benchmarks to assess the effectiveness of adversarial prompts against AI safety measures.
Parse Score