Mathieu Baudier

Since GGUF is the format used by [llama.cpp](https://github.com/ggerganov/llama.cpp) (which is the most widely used native tool for running models locally), one tends to first rather download the GGUF files in...

Thanks for the details! This sounds perfectly reasonable not to prioritise it. Moreover to have SafeTensor support in Java has the added benefit that the original models tend to be...

> If the private libdir also includes the executables, then using `GGML_BACKEND_DL` together with `GGML_CPU_ALL_VARIANTS` should result in a package with compatibility for most CPU microarchitectures (only supported in x64...

> whisper.cpp should support `GGML_BACKEND_DL` now (added in [ggerganov/whisper.cpp#2843](https://github.com/ggerganov/whisper.cpp/pull/2843)). To close this subtopic, I just wanted to mention that indeed whisper.cpp does not require anymore to build against a ggml-cpu-*...