WebsiteBenchWebsiteBench

Amazon Home Services

Rebuilt by GPT 6.1 Sol: 140 of 799 hidden tests passed (17.5%)

reference.local/
Reference for Amazon Home Services, HomeReference
rebuild.local/
Rebuild by GPT 6.1 Sol, HomeGPT 6.1 Sol
    Every model on Amazon Home Services
    ModelPassed%Screenshot
    GPT 6.1 Sol14017.5
    Claude Opus 5.512415.5
    Grok 4.68710.9
    DeepSeek V4.1 Flash8010.0
    GLM 5.3 Flash455.6
    Claude Opus 5425.3
    Qwen 3.8 Max405.0
    Kimi K3283.5