]> git.djapps.eu Git - pkg/ggml/sources/llama.cpp/commit
overview / pkg / ggml / sources / llama.cpp / commit
summary | shortlog | log | commit | commitdiff | tree
(parent: 7747553)
llama-bench: add --n-cpu-moe support (#15952)
authorjacekpoplawski <redacted>
Tue, 16 Sep 2025 14:17:08 +0000 (16:17 +0200)
committerGitHub <redacted>
Tue, 16 Sep 2025 14:17:08 +0000 (16:17 +0200)
commit8ff206097c2bf3ca1c7aa95f9d6db779fc7bdd68
tree1640e1abef6d24e072a12160ef0dce5787b9fd04tree
parent77475530b8bbea3bf578632507e1284cdfe2c8c0commit | diff
llama-bench: add --n-cpu-moe support (#15952)

* llama-bench: add --n-cpu-moe support

Support --n-cpu-moe in llama-bench the same way it is supported by
llama-server.
common/arg.cpp diff | blob | history
common/common.h diff | blob | history
tools/llama-bench/llama-bench.cpp diff | blob | history
Packaging of ggml-org/llama.cpp