Reasoning prefills on a few open models

By wsxiaoys · 2026-08-16 · 2 points · 0 comments

https://gist.github.com/wsxiaoys/102e8654c14d5d27b7b77532026ebfa5

I wrapped this small followup to stolen-thoughts.com in a gist for easier reading and wanted to share it here. My hunch is that this might not just be reasoning distillation; it could even be benchmark distillation. This is a small experiment and definitely doesn’t prove anythin…

Open the full discussion on BetterNews