
LLMs
LFM2.5-8B-A1B: A More Capable On‑Device Mixture‑of‑Experts Model
5/29/2026

LLMs
MiniMax’s M3 LLM promises faster sparse attention but still faces open questions
5/28/2026

Machine Learning
Fast KV Compaction via Attention Matching: A New Approach to Long Context Scaling
2/20/2026

AI
Recursive Language Models: The Architecture That Actually Solves Long Context Problems
2/8/2026

AI
StepFun Releases Step 3.5 Flash, an Open-Source Foundation Model Built for AI Agents
2/2/2026

LLMs
MIT's Recursive Language Models: A Strategic Solution for Enterprise Long-Context AI Challenges
1/21/2026