Hackernews posts about Johnny LLM

1. 01
   

   newSOTA on the hardest AI memory benchmark (BEAM, 10M tokens), with a smaller model

   ▲ 3johnnymakes68d💬 3savesaved
2. 02
   

   newLLMs' – Failure Modes and Proposed Improvementsgithub.com

   ▲ 1professor_jonny133d💬 1savesaved
3. 03
   

   newShow HN: Mighty – let AI agents use private data safelywww.mightynetwork.ai

   ▲ 1jodoking467ddiscusssavesaved

· edit