WebsiteBenchWebsiteBench

Etsy

Rebuilt by GPT 6.1 Sol: 138 of 347 hidden tests passed (39.8%)

reference.local/
Reference for Etsy, HomeReference
rebuild.local/
Rebuild by GPT 6.1 Sol, HomeGPT 6.1 Sol
    Every model on Etsy
    ModelPassed%Screenshot
    GPT 6.1 Sol13839.8
    Claude Opus 5.513639.2
    Claude Opus 513338.3
    DeepSeek V4.1 Flash9226.5
    Grok 4.69126.2
    GLM 5.3 Flash8925.6—
    Kimi K38223.6
    Qwen 3.8 Max00.0—