diff options
| author | slaren <slarengh@gmail.com> | 2024-03-26 01:16:01 +0100 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-03-26 01:16:01 +0100 |
| commit | 280345968dabc00d212d43e31145f5c9961a7604 (patch) | |
| tree | 4d0ada8b59a4c15cb6d4fe1a6b4740a30dcdb0f2 /examples/imatrix/README.md | |
| parent | b06c16ef9f81d84da520232c125d4d8a1d273736 (diff) | |
cuda : rename build flag to LLAMA_CUDA (#6299)
Diffstat (limited to 'examples/imatrix/README.md')
| -rw-r--r-- | examples/imatrix/README.md | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/examples/imatrix/README.md b/examples/imatrix/README.md index 578e8fc2..458c01b8 100644 --- a/examples/imatrix/README.md +++ b/examples/imatrix/README.md @@ -22,7 +22,7 @@ For faster computation, make sure to use GPU offloading via the `-ngl` argument ## Example ```bash -LLAMA_CUBLAS=1 make -j +LLAMA_CUDA=1 make -j # generate importance matrix (imatrix.dat) ./imatrix -m ggml-model-f16.gguf -f train-data.txt -ngl 99 |