WebsiteBenchWebsiteBench

Blue Apron

Rebuilt by Claude Opus 5.5: 89 of 410 hidden tests passed (21.7%)

reference.local/
Reference for Blue Apron, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Blue Apron
    ModelPassed%Screenshot
    Claude Opus 5.58921.7
    Kimi K36215.1
    Grok 4.65814.1
    GLM 5.3 Flash5312.9
    Claude Opus 54912.0
    GPT 6.1 Sol4110.0
    DeepSeek V4.1 Flash266.3
    Qwen 3.8 Max204.9