BLAST RADIUS

Open-weight & owned stack · Quantization efficiency

llama.cpp Adds Direct-Mapped DMA Cache

Sep 23, 2026, first seen via Training & inference tooling releases

llama.cpp release b11118 added a direct-mapped DMA cache for HVX attention-mask handling and builds across multiple accelerator platforms.

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

More in Open-weight & owned stack