firecoperana 49979ba9e9 llama: enable K-shift for quantized KV cache for cuda (#760) ... cuda: add q8_0->f32 cpy operation (#9571) It will fail on unsupported backends or quant types. Co-authored-by: Ivan <nekotekina@gmail.com>		2025-09-05 11:54:18 +02:00
..
cmake	Merge mainline llama.cpp (#3)	2024-07-27 07:55:01 +02:00
include	Offload only activated experts to the GPU (#698)	2025-09-04 12:22:30 +02:00
src	llama: enable K-shift for quantized KV cache for cuda (#760)	2025-09-05 11:54:18 +02:00
.gitignore	Merge mainline llama.cpp (#3)	2024-07-27 07:55:01 +02:00
CMakeLists.txt	Set default value of GGML_SCHED_MAX_COPIES to 1 (#751)	2025-09-02 07:04:39 +02:00