Brian Lovin
/
Hacker News
Show HN: AutoThink – Boosts local LLM performance with adaptive reasoning