OpenAI's Post-Training Questioned: Fable Model Praised for Superior Taste

willdepue · x · 2026-08-07

The author argues that OpenAI needs a 'code red' for its post-training methods, claiming their models lack the refined taste and thoughtful output seen elsewhere. They specifically praise the 'Fable' model for being non-slop and exceptionally good at naming.

Original post →

More from Models

Models channel →