WebsiteBenchWebsiteBench

Amazon

Rebuilt by Claude Opus 5.5: 582 of 1,025 hidden tests passed (56.8%)

reference.local/
Reference for Amazon, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Amazon
    ModelPassed%Screenshot
    Claude Opus 5.558256.8
    GPT 6.1 Sol49548.3
    Claude Opus 542941.9
    Qwen 3.8 Max40239.2
    GLM 5.3 Flash39838.8
    Kimi K338637.7
    Grok 4.631330.5
    DeepSeek V4.1 FlashError—