File tree Expand file tree Collapse file tree 1 file changed +0
-24
lines changed Expand file tree Collapse file tree 1 file changed +0
-24
lines changed Original file line number Diff line number Diff line change @@ -9,30 +9,6 @@ cd llama.cpp
99
1010In order to build llama.cpp you have four different options.
1111
12- - Using ` make ` :
13- - On Linux or MacOS:
14-
15- ``` bash
16- make
17- ```
18-
19- - On Windows (x86/x64 only, arm64 requires cmake):
20-
21- 1. Download the latest fortran version of [w64devkit](https://github.com/skeeto/w64devkit/releases).
22- 2. Extract ` w64devkit` on your pc.
23- 3. Run ` w64devkit.exe` .
24- 4. Use the ` cd` command to reach the ` llama.cpp` folder.
25- 5. From here you can run:
26- ` ` ` bash
27- make
28- ` ` `
29-
30- - Notes:
31- - For ` Q4_0_4_4` quantization type build, add the ` GGML_NO_LLAMAFILE=1` flag. For example, use ` make GGML_NO_LLAMAFILE=1` .
32- - For faster compilation, add the ` -j` argument to run multiple jobs in parallel. For example, ` make -j 8` will run 8 jobs in parallel.
33- - For faster repeated compilation, install [ccache](https://ccache.dev/).
34- - For debug builds, run ` make LLAMA_DEBUG=1`
35-
3612- Using ` CMake ` :
3713
3814 ``` bash
You can’t perform that action at this time.
0 commit comments