Have URMs and UTs been adopted by frontier models?

A Reddit discussion questions whether frontier labs at big tech have adopted Universal Reasoning Model (URM) and Universal Transformer (UT) enhancements, as UT-based small models trained from scratch outperform standard LLMs on certain tasks by a significant margin.

• EngramEdit Enables Decoupled LLM Knowledge Updates2026-10-09
• 825k‑parameter model generates RP2040‑executable drawing programs2026-09-13