Update app.py

app.py

@@ -4,7 +4,7 @@ from transformers import AutoModel, AutoTokenizer
 import gradio as gr
 from PIL import Image
 import traceback
-import numpy
+import spaces
 # Argument Parsing
 parser = argparse.ArgumentParser(description='demo')
 parser.add_argument('--device', type=str, default='cuda', help='cuda or mps')
@@ -14,11 +14,18 @@ assert device in ['cuda', 'mps']
 
 # Load model and tokenizer manually
 model_path = 'openbmb/MiniCPM-Llama3-V-2_5'
-model = AutoModel.from_pretrained(model_path, trust_remote_code=True, torch_dtype=torch.float16).to(device=device)
+if 'int4' in model_path:
+    if device == 'mps':
+        print('Error: running int4 model with bitsandbytes on Mac is not supported right now.')
+        exit()
+    model = AutoModel.from_pretrained(model_path, trust_remote_code=True)
+else:
+    model = AutoModel.from_pretrained(model_path, trust_remote_code=True).to(dtype=torch.float16)
+    model = model.to(device=device)
 tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
 model.eval()
 
-ERROR_MSG = "Error, please retry"
+
 model_name = 'MiniCPM-Llama3-V 2.5'
 
 def chat(image, msgs, sampling=True, temperature=0.7, stream=False):