vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Update information about BLAS

Browse source
This commit is contained in:
DaniAndTheWeb 2023-04-26 00:38:40 +02:00 • committed by GitHub
parent ab07da07c1
commit e1b704b44c
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

6
README.md
View file

@ -192,11 +192,11 @@ Note: For Windows, CMake or Zig can be used.
zig build -Drelease-fast
```
BLAS acceleration is not enabled by default, there are currently two different implementations of it:
BLAS is not enabled by default but by enabling it you may obtain some performance improvements. There are currently two different implementations of it:
- OpenBLAS
This provides a faster evaluation time using only the CPU. Make sure to have OpenBLAS installed on your machine.
This provides BLAS acceleration using only the CPU. Make sure to have OpenBLAS installed on your machine.
- Using `make`
```bash
@ -215,7 +215,7 @@ BLAS acceleration is not enabled by default, there are currently two different i
```
- cuBLAS
This provides a faster evaluation time using the CUDA cores of your Nvidia GPU. Make sure to have the CUDA toolkit installed.
This provides BLAS acceleration using the CUDA cores of your Nvidia GPU. Make sure to have the CUDA toolkit installed.
- Using `make`
```bash
make LLAMA_CUBLAS=1