Skip to main content
Askalante
  • Source
  • Archive
  • Tags
  • Family

[X.com] by @deedydas

Mac

2025-07-15

Source

Original site

Google DeepMind just dropped this new LLM model architecture called Mixture-of-Recursions.

It gets 2x inference speed, reduced training FLOPs and ~50% reduced KV cache memory. Really interesting read.

Has potential to be a Transformers killer. https://t.co/LdrKmSy6tR

Image from tweet


View original on X.com

  • twitter-archive
  • Previous post
  • Next post
Contents © 2025 Mac - Powered by Nikola