llama.cpp

Block interleaving support for Q4_K quantization for x86 AVX2 architecture

#12332

Merged