Update app.py

app.py

@@ -1,4 +1,4 @@
-from transformers import MllamaForConditionalGeneration, AutoProcessor, TextIteratorStreamer , AutoModel
+from transformers import MllamaForConditionalGeneration, AutoProcessor, TextIteratorStreamer , AutoModel,Qwen2VLForConditionalGeneration
 from qwen_vl_utils import process_vision_info
 from PIL import Image
 import requests
@@ -19,7 +19,7 @@ logger = logging.getLogger(__name__)
 
 # Load model and processor
 ckpt ="Qwen/Qwen2-VL-72B-Instruct-AWQ"
-model = AutoModel.from_pretrained(ckpt, torch_dtype=torch.bfloat16,trust_remote_code=True).to("cuda")
+model = Qwen2VLForConditionalGeneration.from_pretrained(ckpt, torch_dtype=torch.bfloat16,trust_remote_code=True).to("cuda")
 processor = AutoProcessor.from_pretrained(ckpt,trust_remote_code=True)
 
 class DocumentState:
@@ -125,7 +125,7 @@ def process_uploaded_file(file):
         logger.error(f"Error in process_file: {str(e)}")
         return "An error occurred while processing the file. Please try again."
 
-@spaces.GPU()
+@spaces.GPU(duration=660)
 def bot_streaming(prompt_option, max_new_tokens=8192):
     try:
         # Define predetermined prompts