char-fork/llama.cpp

git clone https://git.t4t.associates/char-fork/llama.cpp

Johannes GäßlerCUDA: MMQ support for iq4_nl, iq4_xs (#8278)8e558309d

char/aliceai-t5-moe
140 B5 linesraw
1// This file has been autogenerated by generate_cu_files.py, do not edit manually.
2
3#include "../mmq.cuh"
4
5DECL_MMQ_CASE(GGML_TYPE_IQ4_XS);