FEATURE: Any plans to Linux and only CPU no GPU run?

[bi4key]

Any plans to Linux and only CPU no GPU run?

[Anbeeld]

Linux CPU-only should already work through the normal llama.cpp CPU backend. Build without the CUDA flags:

cmake -B build -DCMAKE_BUILD_TYPE=Release
cmake --build build --config Release -j

Then run e.g.:

./build/bin/llama-server -m model.gguf --n-gpu-layers 0

BeeLlama’s main fork-specific speedups are GPU-oriented, especially DFlash/TurboQuant/TCQ CUDA paths, so I’m not currently planning separate CPU-only release binaries. If you hit a CPU-only Linux build/runtime failure in this fork, please open it with the exact commit, OS/compiler, CMake command, and error log.