llama.cpp

ggml-quants: Provide ggml_vqtbl1q_u8 for 64bit compatibility

#5711

Merged