WebsiteBenchWebsiteBench

Eat This Much

Rebuilt by Claude Opus 5.5: 282 of 668 hidden tests passed (42.2%)

reference.local/
Reference for Eat This Much, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Eat This Much
    ModelPassed%Screenshot
    Claude Opus 5.528242.2
    DeepSeek V4.1 Flash16725.0
    Claude Opus 515723.5
    GPT 6.1 Sol12518.7
    Grok 4.612218.3
    Kimi K3284.2
    Qwen 3.8 Max192.8
    GLM 5.3 Flash172.5