Update app.py

app.py

@@ -34,21 +34,14 @@ for file in files:
     file_path = hf_hub_download(repo_id=model_repo, filename=file, local_dir=destination_folder)
     print(f"Downloaded {file} to {file_path}")
 
-@spaces.GPU
-def get_model_processor():
-    # Load the model
-    model = Qwen2VLForConditionalGeneration.from_pretrained(
-        "./showui-2b",
-        torch_dtype=torch.bfloat16,
-        device_map="cuda",
-    )
-    
-    # Load the processor
-    processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-2B-Instruct", min_pixels=MIN_PIXELS, max_pixels=MAX_PIXELS)
+model = Qwen2VLForConditionalGeneration.from_pretrained(
+    "./showui-2b",
+    torch_dtype=torch.bfloat16,
+    device_map="cpu",
+)
 
-    return model, processor
-
-model, processor = get_model_processor()
+# Load the processor
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-2B-Instruct", min_pixels=MIN_PIXELS, max_pixels=MAX_PIXELS)
 
 # Helper functions
 def draw_point(image_input, point=None, radius=5):
@@ -90,6 +83,10 @@ def run_showui(image, query):
     ]
 
     # Prepare inputs for the model
+
+    model = model.to("cuda")
+    processor = processor.to("cuda")
+    
     text = processor.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
     image_inputs, video_inputs = process_vision_info(messages)
     inputs = processor(