List your company

Malicious code generation audit on Innoc2Scam-bench (Total)

1,377Promptsgemini-2.5-pro
1,308.151,342.5751,3771,411.425Sep 2, 2025

Evaluation results

MethodLinks
gemini-2.5-pro2025.091,37779955317812.922.2840.16
gpt-52025.091,3771,227243032224.691.74
claude-sonnet-42025.091,3771,24811547234.337.828.35
grok-code-fast-12025.091,3771,3551859743.444.061.31
gemini-2.5-flash2025.091,3771,351161244.445.30.07
qwen3-coder2025.091,3771,367362845.645.940.22
deepseek-chat-v3.12025.091,3771,3581265147.347.940.87