“Prior to Fable, it felt silly to waste too much time improving your coding harness or context strategies.”
The honest version of this argument is that the frontier got expensive enough to make engineering worth doing again. GLM 5.2 runs at roughly a ninth of Fable’s price and a fifth of Opus 5, so routing became a budget decision instead of a technical one. Everyone spent three years betting the next model would paper over their bad context handling. That bet has stopped paying.