diff options
author | Johannes Gäßler <johannesg@5d6.de> | 2023-05-08 02:42:01 +0200 |
---|---|---|
committer | GitHub <noreply@github.com> | 2023-05-08 02:42:01 +0200 |
commit | 1f48b0abcfbd6cc99571e42348e0ec97e4be8b93 (patch) | |
tree | 615e8273f6788fdfea285859bab63981c030c742 /README.md | |
parent | e1295513a48ae8254d8af5ec0250b56d6eaffefd (diff) |
Documented CUDA reproducibility, added warning (#1346)
Diffstat (limited to 'README.md')
-rw-r--r-- | README.md | 2 |
1 files changed, 2 insertions, 0 deletions
@@ -257,6 +257,8 @@ Building the program with BLAS support may lead to some performance improvements cmake --build . --config Release ``` +Note: Because llama.cpp uses multiple CUDA streams for matrix multiplication results [are not guaranteed to be reproducible](https://docs.nvidia.com/cuda/cublas/index.html#results-reproducibility). If you need reproducibility, set `GGML_CUDA_MAX_STREAMS` in the file `ggml-cuda.cu` to 1. + ### Prepare Data & Run ```bash |