fix

Abdennacer-Badaoui · Abdennacer-Badaoui · commit c72f9e808593 · 2026-02-20T16:39:59.000+01:00
diff --git a/docs/source/quickstart.mdx b/docs/source/quickstart.mdx
@@ -122,6 +122,7 @@ for batch in dataloader:
 Use quantized linear layers directly in your models:
 
 ```py
+import torch
 import bitsandbytes as bnb
 
 # 8-bit linear layer
