WebsiteBenchWebsiteBench

Fandango

Rebuilt by GPT 6.1 Sol: 43 of 387 hidden tests passed (11.1%)

reference.local/
Reference for Fandango, HomeReference
rebuild.local/
Rebuild by GPT 6.1 Sol, HomeGPT 6.1 Sol
    Every model on Fandango
    ModelPassed%Screenshot
    GPT 6.1 Sol4311.1
    Claude Opus 5.54210.9
    Kimi K34210.9
    Claude Opus 5338.5
    Qwen 3.8 Max256.5
    Grok 4.692.3
    GLM 5.3 Flash71.8
    DeepSeek V4.1 FlashError—

    Every result in the WaffleRun the benchmark