Update app.py

app.py

@@ -4,7 +4,7 @@ from threading import Thread
 import re
 import time
 import torch
-import spaces
+#import spaces
 #import subprocess
 #subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
@@ -16,7 +16,7 @@ model = AutoModelForVision2Seq.from_pretrained("HuggingFaceTB/SmolVLM2-500M-Inst
                                                torch_dtype=torch.bfloat16).to("cuda:0")
 
 
-@spaces.GPU
+#@spaces.GPU
 def model_inference(
     input_dict, history, max_tokens
 ):