I love this project:
Initial low-rank adaptation support has been added to llama.cpp We now have the option to apply LoRA adapters to a base model at runtime. Lots of room for improvements and opens up possibilities
I love this project:
Initial low-rank adaptation support has been added to llama.cpp We now have the option to apply LoRA adapters to a base model at runtime. Lots of room for improvements and opens up possibilities