Pro tip: raise reasoning effort to stop GPT-6.2 Sol quitting early on long-horizon tasks

pvncher · x · 2026-11-18

pvncher shares a practical tip for long-horizon work with GPT-6.2 Sol and Astra: if the model stops early, try increasing reasoning effort before reaching for goal mode. He doesn't usually recommend max effort, but for letting the model work for hours, it's worth a shot.

Original post →

More from coding & agent

coding & agent channel →