github ggml-org/llama.cpp b7484

2 hours ago

Warning

Release Format Update: Linux releases will soon use .tar.gz archives instead of .zip. Please make the necessary changes to your deployment scripts.

Details

ggml-hexagon: Implement true Q8_0 quantization on Hexagon NPU for more accurate mixed-precision matmul operations (#17977)

  • feat: implement real Q8_0

  • feat: adding cmake option for configuring FP32 quantize group size

  • typo: set() shall be used


Co-authored-by: ngdxzy zhenyu_xu@uri.edu

macOS/iOS:

Linux:

Windows:

openEuler:

Don't miss a new llama.cpp release

NewReleases is sending notifications on new releases.