List your company

Agent Decision Making on 50-storefront golden set skimmers cohort (overall)

77Alignment (%)Gemini 3 Flash
58.2863.146872.86May 19, 2026

Evaluation results

MethodLinks
Gemini 3 FlashInput Modality=Visual, Runtime/shop=5.3 min2026.0577660.550.32
Gemini 3 Flash (text only)Input Modality=Text-only, Runtime/shop=4.5 min2026.0570580.490.25
GPT-OSS (text only)Input Modality=Text-only, Runtime/shop=12.8 min2026.0559470.410.15