WebsiteBenchWebsiteBench

Capterra

Rebuilt by Claude Opus 5.5: 241 of 632 hidden tests passed (38.1%)

reference.local/
Reference for Capterra, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Capterra
    ModelPassed%Screenshot
    Claude Opus 5.524138.1
    GPT 6.1 Sol19831.3
    Claude Opus 519230.4
    GLM 5.3 Flash17527.7
    Kimi K316826.6
    DeepSeek V4.1 Flash16526.1
    Grok 4.6172.7
    Qwen 3.8 Max152.4