the more i hear ppl complaining that 4.5 isn't beating prev models on evals the more excited i get abt it bec must be better in other ways- OOD tasks, creativity, EQ, soul... https://x.com/aidan_mclau/status/1895204304056897934
mimi10v3mimi10v3mimi10v3the more i hear ppl complaining that 4.5 isn't beating prev models on evals the more excited i get abt it bec must be better in other ways- OOD tasks, creativity, EQ, soul... https://x.com/aidan_mclau/status/1895204304056897934