]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 19f68fa)
llama : add --n-cpu-moe option (#15077)
authorDiego Devesa <redacted>
Mon, 4 Aug 2025 23:05:36 +0000 (16:05 -0700)
committerGitHub <redacted>
Mon, 4 Aug 2025 23:05:36 +0000 (01:05 +0200)
commitec428b02c347767f24c78111309e3f30d2ada289
tree8189bb9f07c2ceecba5d3f74238105454b801b80tree
parent19f68fa5a4c3bf796de52a6db9008e77d29f423acommit | diff
llama : add --n-cpu-moe option (#15077)

* llama : add --n-cpu-moe option

Keeps the MoE weights of the first N layers in the CPU
common/arg.cpp diff | blob | history
Packaging of ggml-org/llama.cpp