Skip to content

Open source / #14772

Open source

ggml-org/llama.cpp b10643

GitHub Releases · github-actions[bot] · 22 days ago

<details open> hexagon: support for multi-NPU devices (IQ9, IQ10) and fully asynchronous backend (#26501) * hexagon: use non-host bufs by default and make the backend fully async * hex-hb: remove optional hostbuf support and fix async copy * hex-unary: relax supported unary chec…

Source adapter
github-releases
Source weight
3
Points
0
Age
536.7 h
Stored score
0.00002
Current score
0.00002
First seen
2026-08-27T08:37:06.035Z

Newsletter

Get practical AI engineering notes

Receive source-checked analysis of models, agents, evaluation, retrieval, and production reliability. Sent only when there is useful work to share.