]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 7eb4117)
CUDA: use only 1 thread if fully offloaded (#2915)
authorJohannes Gäßler <redacted>
Thu, 21 Sep 2023 08:43:53 +0000 (10:43 +0200)
committerGitHub <redacted>
Thu, 21 Sep 2023 08:43:53 +0000 (11:43 +0300)
commit8185710a80531e9ee0c0cb99d3a9c9af1019ab67
tree1486dc344c960bf215581b111bd66cee69a1396btree
parent7eb41179edc56083ef4eb2df7967ac9ff38b34fbcommit | diff
CUDA: use only 1 thread if fully offloaded (#2915)
llama.cpp diff | blob | history
Packaging of ggml-org/llama.cpp