WebsiteBenchWebsiteBench

Eventbrite

Rebuilt by Claude Opus 5.5: 376 of 1,655 hidden tests passed (22.7%)

reference.local/
Reference for Eventbrite, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Eventbrite
    ModelPassed%Screenshot
    Claude Opus 5.537622.7
    GPT 6.1 Sol1458.8
    DeepSeek V4.1 Flash764.6
    Grok 4.6472.8
    GLM 5.3 Flash432.6
    Claude Opus 5422.5
    Qwen 3.8 Max412.5
    Kimi K3392.4