Repository navigation

build

FoldQuant W8A8/W4A4 inference and FoldQuantVLA quantized-model converter #83

Job	Run time
build-cuda	2m 54s
py-tooling	4s
build-gate	2m 38s
build-backends (metal, macos-15, -DGGML_METAL=ON, true)	1m 40s
build-backends (opencl, ubuntu-24.04, -DGGML_OPENCL=ON -DVLA_WERROR=ON)	1m 18s
	8m 34s