Update app.py

app.py

@@ -35,7 +35,7 @@ def array_to_image_path(image_array):
     return full_path
     
 models = {
-    "Qwen/Qwen2-VL-7B-Instruct": Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", trust_remote_code=True, torch_dtype="auto").cuda().eval()
+    "Qwen/Qwen2-VL-7B-Instruct": Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", trust_remote_code=True, torch_dtype=torch.float16,device_map='auto')
 
 }
 
@@ -52,7 +52,7 @@ user_prompt = '<|user|>\n'
 assistant_prompt = '<|assistant|>\n'
 prompt_suffix = "<|end|>\n"
 
-@spaces.GPU
+# @spaces.GPU
 def run_example(image, text_input=None, model_id="Qwen/Qwen2-VL-7B-Instruct"):
     image_path = array_to_image_path(image)
     
@@ -87,8 +87,7 @@ def run_example(image, text_input=None, model_id="Qwen/Qwen2-VL-7B-Instruct"):
         padding=True,
         return_tensors="pt",
     )
-    inputs = inputs.to("cuda")
-    
+    # inputs = inputs.to("cuda")
     # Inference: Generation of the output
     generated_ids = model.generate(**inputs, max_new_tokens=1024)
     generated_ids_trimmed = [