Comment on Ask HN: What will happen as AI costs increase?parentComments−yulaow4moopus 4.6 -> 4.7 did not respect this assumption−atleastoptimal4mo4.7 is better imo and uses fewer tokens per requests−brazukadev4moand it did not even got better.
Comments
opus 4.6 -> 4.7 did not respect this assumption
4.7 is better imo and uses fewer tokens per requests
and it did not even got better.