BLAST RADIUS

Open-weight & owned stack · Quantization efficiency

llama.cpp Adds Model and Runtime Improvements

Sep 14, 2026, first seen via Training & inference tooling releases

llama.cpp released version 0.4.1 with support for three models and improvements to recurrent state, KV-cache, JSON, conversion, and server handling.

Read the original at github.comOpens the publisher's site in a new tab

More in Open-weight & owned stack

Xiaomi Publishes MiMo-V2.6 Reinforcement Learning DashboardSep 17Gemma Models Surpass One Billion DownloadsSep 16llama.cpp Fixes Remote-Code-Execution VulnerabilitySep 16Alibaba Releases Occamy Open-Weight AI AgentSep 15Apple Enables Local Foundation Models On MacOSSep 15