>>106166542 (OP)Yes, the same way you can get VEO 3, put in anime images and make your own anime show. Both will be a slop with disconnected events happening, both will have hallucinated details, both will forget what happened few minutes ago, both will give you the most sloppish basic dialogue and story unless you put your own dialogue into the prompt and both will cost entire forest worth of energy to generate that slop. They aren't talking about this as a text2game model, they are talking about this as a world model, basically you walking into a video and maybe doing some interactions.