Update app.py

app.py

@@ -32,19 +32,19 @@ import torch
 
 
 
-# Create a configuration for quantization
-quantization_config = BitsAndBytesConfig(
-    load_in_4bit=True,   
-    bnb_4bit_compute_dtype="float16",  
-    bnb_4bit_use_double_quant=True,    
-    bnb_4bit_quant_type="nf4",         
-)
+# # Create a configuration for quantization
+# quantization_config = BitsAndBytesConfig(
+#     load_in_4bit=True,   
+#     bnb_4bit_compute_dtype="float16",  
+#     bnb_4bit_use_double_quant=True,    
+#     bnb_4bit_quant_type="nf4",         
+# )
 
 # Load the model and processor
 model = Qwen2VLForConditionalGeneration.from_pretrained(
     "Qwen/QVQ-72B-Preview", device_map="auto", 
     quantization_config=quantization_config,
-    offload_folder="offload",
+    # offload_folder="offload",
 )
 processor = AutoProcessor.from_pretrained("Qwen/QVQ-72B-Preview")
 
@@ -58,7 +58,8 @@ def process_image_and_question(image, question):
         {
             "role": "system",
             "content": [
-                {"type": "text", "text": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step."}
+                # {"type": "text", "text": "You are a helpful and harmless assistant. You are Qwen developed by Alibaba. You should think step-by-step."}
+                {"type": "text", "text": "You are helpful assistant, you give answer in JSON"}
             ],
         },
         {