jevIMAGE
Waldo benchmark: Jev 1.14 fastest, d1-ultra finds the most
Across 66 scenes of a Where's Waldo benchmark, Jev 1.14 was the quickest decision model, LiquidAI d1-ultra found the most Waldos, and Perplexity Decider Flash V1.2 was cheapest.
Andy Gao
@instantricecookArchived 11 days agoOriginally posted 2026.10.19
Andy Gao @instantricecookOriginally posted 2026.10.19
I wanted to see how well decision models are at... Finding Waldo. After 66 scenes, here are the results: - @liquidai d1-ultra found the most Waldos - @typesafeai Jev 1.14 was the quickest -@perplexity_ai Decider Flash V1.2 was the cheapest https://waldobench.gao.haus
x.com/instantricecook/status/2111971372888998305Archived 2026.10.19 10:00