Comment by rw2

Totally agree with all of this. Most of the problems we solve with AI are not one shot, all the bench marks also miss the human components when testing. Thinking alongside with a human and coming to a solution is what Fable does better than most other models.