Dev trains a tiny 2B model just for FunSearch, then gives it a shell tool
cephaloform · x · 2026-11-21
Developer @cephaloform shows off a 2B model trained specifically for FunSearch — an early auto-research harness that even works on base models — now equipped with tools including shell access. The takeaway: an autonomous research loop can be squeezed into a tiny specialized model instead of relying on frontier-size LLMs.
More from coding & agent
- Sebastian Raschka: the harness is part of training, LLMs can't absorb it — rasbt · 2026-11-21
- Astra 6.2 generates an ocean simulation with buoyant balls in one HTML file — mark_k · 2026-11-21
- LangChain: scheduled runs hit 84% of agent runs after five weeks deployed — LangChain · 2026-11-21
- Dev insight: the most important part of agent harnesses is stopping AI shortcuts — clemnt · 2026-11-21
- Jeremy Howard: tying Neon to Antigravity will limit its success — jeremyphoward · 2026-11-21
- World's first AI Community Note writing competition open, $10k prizes, entries due Dec 8 — NathanpmYoung · 2026-11-21