diff --git a/scripts/convert_original_stable_diffusion_to_diffusers.py b/scripts/convert_original_stable_diffusion_to_diffusers.py
index 0414a0e8ad6a..1f4204495482 100644
--- a/scripts/convert_original_stable_diffusion_to_diffusers.py
+++ b/scripts/convert_original_stable_diffusion_to_diffusers.py
@@ -848,12 +848,17 @@ def convert_open_clip_checkpoint(checkpoint):
         ),
     )
     parser.add_argument("--dump_path", default=None, type=str, required=True, help="Path to the output model.")
+    parser.add_argument("--device", type=str, help="Device to use (e.g. cpu, cuda:0, cuda:1, etc.)")
     args = parser.parse_args()
 
     image_size = args.image_size
     prediction_type = args.prediction_type
 
-    checkpoint = torch.load(args.checkpoint_path)
+    if args.device is None:
+        device = "cuda" if torch.cuda.is_available() else "cpu"
+        checkpoint = torch.load(args.checkpoint_path, map_location=device)
+    else:
+        checkpoint = torch.load(args.checkpoint_path, map_location=args.device)
 
     # Sometimes models don't have the global_step item
     if "global_step" in checkpoint: