diff options
| author | William Tambellini <william.tambellini@gmail.com> | 2024-05-06 11:12:14 -0700 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-05-06 20:12:14 +0200 |
| commit | 858f6b73f6e57a62523d16a955d565254be889b4 (patch) | |
| tree | f6a64462f2173d18a986ab1b58a8ef746869bfbb /common/sampling.h | |
| parent | b3a995b416e13ae3123a117a743e11d0ede0ca4c (diff) | |
Add an option to build without CUDA VMM (#7067)
Add an option to build ggml cuda without CUDA VMM
resolves
https://github.com/ggerganov/llama.cpp/issues/6889
https://forums.developer.nvidia.com/t/potential-nvshmem-allocated-memory-performance-issue/275416/4
Diffstat (limited to 'common/sampling.h')
0 files changed, 0 insertions, 0 deletions