Michael Yang a00fac4ec8 update llama.cpp

2023-11-21 09:50:02 -08:00

..

0001-add-detokenize-endpoint.patch

…

0001-copy-cuda-runtime-libraries.patch

…

0001-update-default-log-target.patch

update default log target

2023-10-23 10:44:50 -07:00

0002-34B-model-support.patch

…

0003-metal-fix-synchronization-in-new-matrix-multiplicati.patch

…

0004-metal-add-missing-barriers-for-mul-mat-2699.patch

…

0005-ggml-support-CUDA-s-half-type-for-aarch64-1455-2670.patch

…