✨ Sparkfile
A hobbyist build of mozilla-ai/llamafile for DGX Spark.
v0.10.0
both CUDA and Vulkan backends packaged.
after download validate backends
chmod +x ./sparkfile
./sparkfile --gpu nvidia --cli -m your_local_model.gguf --list-devices
./sparkfile --gpu vulkan --cli -m your_local_model.gguf --list-devices
*should see both CUDA and Vulkan backends being reported.
basic usage
./sparkfile --chat -m your_local_model.gguf
./sparkfile --gpu nvidia --chat -m your_local_model.gguf
./sparkfile --gpu vulkan --chat -m your_local_model.gguf
./sparkfile --server ... <llama-server args>
for certain reason you are not launching as an executable itself, e.g. wsl2
sh ./sparkfile <args>
Expected Platforms and OSs CPU: ANY if supported by llamafile GPU:
- aarch64_cpu_linux: DGX Spark (GB10), Jetson Thor
- x86-64_cpu_linux: (needs a seperate
.soor recompile) - x86-64_cpu_win: (needs a seperate
.dllor recompile)