Update app.py

app.py

@@ -12,6 +12,7 @@ from llava.constants import IMAGE_TOKEN_INDEX, DEFAULT_IMAGE_TOKEN
 from llava.model.builder import load_pretrained_model
 from llava.mm_utils import tokenizer_image_token
 from transformers.generation.streamers import TextIteratorStreamer
+import spaces
 
 def load_image(image_file):
     if image_file.startswith('http') or image_file.startswith('https'):
@@ -21,6 +22,7 @@ def load_image(image_file):
         image = Image.open(image_file).convert('RGB')
     return image
 
+@spaces.GPU
 def bot_streaming(message, history):
     device = "cuda:0"
     tokenizer, model, image_processor, context_len = load_pretrained_model(model_path="./checkpoints/llava-phi3-3.8b-lora", model_name="llava-phi3-3.8b-lora", model_base="microsoft/Phi-3-mini-128k-instruct", load_8bit=False, load_4bit=False, device=device)