llama.cpp
Extend llm_build_ffn() to support _scale tensors
#8103
Merged

Loading