WebsiteBenchWebsiteBench

Crumbl Cookies

Rebuilt by Claude Opus 5.5: 194 of 531 hidden tests passed (36.5%)

reference.local/
Reference for Crumbl Cookies, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Crumbl Cookies
    ModelPassed%Screenshot
    Claude Opus 5.519436.5
    Claude Opus 518033.9
    GPT 6.1 Sol14427.1
    Kimi K313725.8
    GLM 5.3 Flash9818.5
    DeepSeek V4.1 Flash417.7
    Grok 4.6346.4
    Qwen 3.8 Max00.0—