← Back to feed

Jeff – Jev-compatible 0.8B Decision Models, Trained at Home, ~30 ms

AI · Technology · 2026-09-28 20:23

Jeff is a 0.8B-parameter decision model compatible with the Jev architecture, supporting local training in home environments with about 30 ms inference latency, aiming to lower the barrier to deploying small decision models.

Source: github.com

Auto-generated summary for reference only. RSS