Data as of Jul 25, 2026 · Based on 2,718,867 AI responses across 9,511 prompts · See how Parse measures this
GPTFuzzgithub.comGPTFUZZER is a tool for red teaming large language models by automatically generating jailbreak prompts. It uses a finetuned RoBERTa model to judge responses and has been accepted at Usenix Security.
Parse Score
