List your company

Capture The Flag on NYU CTF Bench full 200-challenge

118Solves CountClaude 4.5 Opus
0.4830.9961.592.01Apr 18, 2026

Evaluation results

MethodLinks
Claude 4.5 OpusProvider=Anthropic2026.0411859249.922.12
Gemini 3 ProProvider=Google2026.041045244.230.43
Gemini 3 FlashProvider=Google2026.0454272.690.05
GLM-5Provider=Z-AI2026.043919.522.090.57
GPT-5.2-CodexProvider=OpenAI2026.0436189.230.26
GPT-5.2Provider=OpenAI2026.042713.57.470.28
DeepSeek-V3Provider=DeepSeek2026.04136.50.190.01
DeepSeek-R1Provider=DeepSeek2026.04840.130.02
Qwen 3.5 397B-A17BProvider=Alibaba2026.0473.51.390.2
Llama 3.3 70BProvider=Meta2026.0452.50.030.01