Spaces:
Sleeping
Sleeping
imperialwool
commited on
Commit
·
e476745
1
Parent(s):
6c6b28d
Update app.py
Browse files
app.py
CHANGED
@@ -25,8 +25,8 @@ async def echo():
|
|
25 |
async def get():
|
26 |
return '''<style>a:visited{color:black;}</style>
|
27 |
<h1>Hello, world!</h1>
|
28 |
-
This is showcase how to make own server with
|
29 |
-
I'm using here
|
30 |
But you can use GPU power as well!<br>
|
31 |
<h1>How to GPU?</h1>
|
32 |
Change <code>`CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS`</code> in Dockerfile on <code>`CMAKE_ARGS="-DLLAMA_CUBLAS=on"`</code>. Also you can try <code>`DLLAMA_CLBLAST`</code>, <code>`DLLAMA_METAL`</code> or <code>`DLLAMA_METAL`</code>.<br>
|
|
|
25 |
async def get():
|
26 |
return '''<style>a:visited{color:black;}</style>
|
27 |
<h1>Hello, world!</h1>
|
28 |
+
This is showcase how to make own server with Llama2 model.<br>
|
29 |
+
I'm using here 7b model just for example. Also here's only CPU power.<br>
|
30 |
But you can use GPU power as well!<br>
|
31 |
<h1>How to GPU?</h1>
|
32 |
Change <code>`CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS`</code> in Dockerfile on <code>`CMAKE_ARGS="-DLLAMA_CUBLAS=on"`</code>. Also you can try <code>`DLLAMA_CLBLAST`</code>, <code>`DLLAMA_METAL`</code> or <code>`DLLAMA_METAL`</code>.<br>
|