Fable uses 3x more tokens than Astra on letter-counting yet scores worse

maksym_andr · x · 2026-09-18

Part 3 of maksymandr's thread: on 1280-character letter-counting, Fable consumes over 3x the tokens of GPT-6-Astra while performing worse, showing Astra achieves higher accuracy with far fewer tokens.

Related event: Letter-counting tests expose Fable 5.1 false safety triggers and poor token efficiency(5 posts)→

Original post →

More from Models

Models channel →