
@diegocabezas01 Cool but since LLMs are probabilistic, shouldn’t you generate like 5 for each model and then see which produces the greatest number of good results? or which produces the best result after X attempts?
English
Stu Denman
1.3K posts

@studenman
Indie game developer & founder of @pinestcw. Previously Surreal Software and Electrolab Games. Currently making #TinyBubbles, an organic puzzle game for humans.


















🦊⚔ Your journey begins today. 🦊⚔ TUNIC is OUT NOW Day One on Xbox Game Pass: Xbox One, Xbox Series X|S, & PC alongside PC and Mac storefronts. What will you discover? tunicgame.com 🛒: microsoft.com/store/r/9nlrt3…