Update app.py

app.py

@@ -6,6 +6,10 @@ import os
 import subprocess
 subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
 
+
+subprocess.run('pip install -U bitsandbytes')
+
+
 # Define the repository for the quantized model
 repo_name = "cyan2k/molmo-7B-D-bnb-4bit"