llama.cpp/scripts/build-cpu.sh
Diego Devesa 3420909dff
ggml : automatic selection of best CPU backend (#10606)
* ggml : automatic selection of best CPU backend

* amx : minor opt

* add GGML_AVX_VNNI to enable avx-vnni, fix checks
2024-12-01 16:12:41 +01:00

13 lines
313 B
Bash
Executable File

#!/bin/bash
name="$1"
args="${@:2}"
echo "Building $name with args: $args"
rm -fr build-cpu-$1
cmake -S . -B build-cpu-$1 -DGGML_BACKEND_DL=ON -DGGML_NATIVE=OFF $args
cmake --build build-cpu-$1 --config Release -t ggml-cpu -j $(nproc)
cp build-cpu-$1/bin/libggml-cpu.so ./libggml-cpu-$1.so
rm -fr build-cpu-$1