Distilling an Agent Down to 2B for Consumer GPUs via DAgger-Style Loops

Kangwook_Lee · x · 2026-09-29

The author shares a full pipeline for shrinking a model to run on consumer GPUs:

The key is a deploy-collect-correct-retrain closed loop rather than one-shot offline distillation.

Related event: Kangwook Lee's Team Rented a Korean PC Bang, Recruited 1000+ Players to Train On-Device Game AI Ally Distilled to 2B(5 posts)→

Original post →

More from coding & agent

coding & agent channel →