Update app.py

app.py

@@ -7,6 +7,7 @@ import traceback
 import re
 import torch
 import argparse
+import numpy as np
 from transformers import AutoModel, AutoTokenizer
 
 # README, How to run demo on different devices
@@ -25,7 +26,7 @@ device = args.device
 assert device in ['cuda', 'mps']
 
 # Load model
-model_path = AutoModel.from_pretrained("openbmb/MiniCPM-Llama3-V-2_5", trust_remote_code=True)
+model_path = AutoModel.from_pretrained("openbmb/MiniCPM-Llama3-V-2_5", trust_remote_code=True,torch_dtype=torch.float32, device_map="auto")
 if 'int4' in model_path:
     if device == 'mps':
         print('Error: running int4 model with bitsandbytes on Mac is not supported right now.')