WebsiteBenchWebsiteBench

Wayfair

Rebuilt by Claude Opus 5.5: 167 of 1,163 hidden tests passed (14.4%)

reference.local/
Reference for Wayfair, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Wayfair
    ModelPassed%Screenshot
    Claude Opus 5.516714.4
    DeepSeek V4.1 Flash766.5
    Grok 4.6655.6
    Claude Opus 5615.2
    GLM 5.3 Flash514.4
    Kimi K3181.5
    Qwen 3.8 Max80.7—
    GPT 6.1 SolError