---

cd ~/llama.cpp
rm -rf build

cmake -B build \
  -DGGML_NATIVE=OFF \
  -DGGML_SSE42=ON \
  -DGGML_AVX=OFF \
  -DGGML_AVX2=OFF \
  -DGGML_FMA=OFF \
  -DGGML_F16C=OFF \
  -DGGML_BMI2=OFF \
  -DCMAKE_C_FLAGS="-mno-avx -mno-avx2 -mno-fma -mno-f16c -mno-bmi2 -mno-sse4.2 -msse4.1" \
  -DCMAKE_CXX_FLAGS="-mno-avx -mno-avx2 -mno-fma -mno-f16c -mno-bmi2 -mno-sse4.2 -msse4.1"

cmake --build build --config Release -j 16

cd ~/llama.cpp
./build/bin/llama-server --version

nohup numactl --cpunodebind=0 --membind=0 ./build/bin/llama-server \
  -m ~/models/RavenX-CyberAgent-35B-v5.1-Q4_K_M.gguf \
  --host 0.0.0.0 --port 8080 \
  -c 8192 -t 8 \
  --cpu-moe --no-mmap --mlock \
  > ~/llama-server.log 2>&1 &


pgrep -a llama-server
tail -10 ~/llama-server.log