Update app.py

app.py

@@ -6,14 +6,15 @@ from PIL import Image
 import torch
 import spaces
 import subprocess
-subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
+#subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
 
 processor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM_converted_4")
 
 model = Idefics3ForConditionalGeneration.from_pretrained("HuggingFaceTB/SmolVLM_converted_4", 
         torch_dtype=torch.bfloat16,
-        _attn_implementation="flash_attention_2").to("cuda")
+        #_attn_implementation="flash_attention_2"
+                                                        ).to("cuda")
 
 @spaces.GPU
 def model_inference(