WebsiteBenchWebsiteBench

Blinkist

Rebuilt by Claude Opus 5.5: 120 of 437 hidden tests passed (27.5%)

reference.local/
Reference for Blinkist, HomeReference
rebuild.local/
Rebuild by Claude Opus 5.5, HomeClaude Opus 5.5
    Every model on Blinkist
    ModelPassed%Screenshot
    Claude Opus 5.512027.5
    GPT 6.1 Sol10123.1
    GLM 5.3 Flash9020.6
    Kimi K38820.1
    Claude Opus 58218.8
    DeepSeek V4.1 Flash7918.1
    Qwen 3.8 Max6815.6
    Grok 4.6439.8