WebsiteBenchWebsiteBench

Curology

Rebuilt by Claude Opus 5.5: 102 of 539 hidden tests passed (18.9%)

reference.local/
Reference for Curology, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Curology
    ModelPassed%Screenshot
    Claude Opus 5.510218.9
    Claude Opus 58515.8
    GPT 6.1 Sol8215.2
    DeepSeek V4.1 Flash7914.7
    GLM 5.3 Flash7614.1
    Grok 4.67213.4
    Kimi K3325.9
    Qwen 3.8 Max00.0—