]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
CUDA: use mul_mat_q kernels by default (#2683)
authorJohannes Gäßler <redacted>
Tue, 22 Aug 2023 20:47:05 +0000 (22:47 +0200)
committerGitHub <redacted>
Tue, 22 Aug 2023 20:47:05 +0000 (22:47 +0200)
commitc63bb1d16a70c03440671b76954bb767513cead8
tree5562dd98e6a37a1bfe635bf25114e3bdb5660996
parent3b6cfe7c927df178ca3c11643c3ec93e143471c9
CUDA: use mul_mat_q kernels by default (#2683)
common/common.cpp
common/common.h
examples/server/server.cpp
ggml-cuda.cu