Open-weight & owned stack · Quantization efficiency

llama.cpp b10909 Fixes Metal Regression

Sep 11, 2026, first seen via Training & inference tooling releases

llama.cpp b10909 reworked Metal fusion tables, fixed a reported token-generation regression, and added cache fusion.

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

More in Open-weight & owned stack

Z.ai publishes GLM-5.3 open weightsSep 8 MiniMax H3 highlighted as open-weight multimodal modelSep 7 InternLM releases 397B multimodal Intern-S2 modelSep 13 DeepSeek Model Nearly Matches GPT-6 Astra DesignSep 10 DeepSeek presents V4 FlashSep 8