# clone llama.cpp repo
git clone https://github.com/ggml-org/llama.cpp

# build llama.cpp
cd llama.cpp
cmake -B build -DGGML_CUDA=ON
cmake --build build --config Release -j

# execute llama.cpp commandline app
/appfolder/app/llama.cpp/build/bin/llama-cli -m /appfolder/mountvolume/Qwen3.6-35B-A3B-UD-Q5_K_XL.gguf

# spin up llama.cpp server
/appfolder/app/llama.cpp/build/bin/llama-server -m /appfolder/mountvolume/Qwen3.6-35B-A3B-UD-Q8_K_XL.gguf



/appfolder/mountvolume/llama.cpp/build/bin/llama-server \
  -m /appfolder/mountvolume/Tiel-Coder-35B-A3B-UD-IQ3_XXS.gguf \
  --host 0.0.0.0 \
  --port 8080 \
  --fit-ctx 131072