WebsiteBenchWebsiteBench

BeerAdvocate

Rebuilt by GPT 6.1 Sol: 183 of 446 hidden tests passed (41.0%)

reference.local/
Reference for BeerAdvocate, HomeReference
rebuild.local/
Rebuild by GPT 6.1 Sol, HomeGPT 6.1 Sol
    Every model on BeerAdvocate
    ModelPassed%Screenshot
    GPT 6.1 Sol18341.0
    GLM 5.3 Flash12127.1
    Claude Opus 5.56414.3
    Grok 4.66414.3
    DeepSeek V4.1 Flash4810.8
    Qwen 3.8 Max4810.8
    Kimi K34710.5
    Claude Opus 5449.9