]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
CUDA: mul_mat_id always on GPU for batches >= 32 (#4553)
authorJohannes Gäßler <redacted>
Thu, 21 Dec 2023 17:42:59 +0000 (18:42 +0100)
committerGitHub <redacted>
Thu, 21 Dec 2023 17:42:59 +0000 (18:42 +0100)
commit9154494808dc865475c59022c29060b4947a803b
tree9e17a6929fa6f704d336476fc8ac17028091f20f
parentc083718c895b7c8c7fb2a4660643fb78d0c64dfd
CUDA: mul_mat_id always on GPU for batches >= 32 (#4553)
ggml-cuda.cu