llama.net
llama.net copied to clipboard
hpretila
.NET wrapper for LLaMA.cpp for LLaMA language model inference on CPU. 🦙
Sorry I intended to contribute to llama.net but i rewrote most of it in the end, so now its unmergeable :D
Would it be possible to change to target framework to 4.8 so it can be used in Unity Engine?
#4 and #2 discussion indicates out-of-sync codebase; sync with latest LLama.CPP. Unsure if model spec will be stable enough to address, ideally pursue #3 for standard quantisation method.
LLaMA should include quantisation. This introduces a dilemma between two options: - The quantisation is done by invoking the current Python interpreter available on the path to convert model state...