Update app.py

app.py

@@ -14,6 +14,7 @@ import spaces
 import gradio as gr
 
 import torch
+torch.backends.cuda.enable_mem_efficient_attention(False)
 import matplotlib.pyplot as plt 
 from PIL import Image, ImageDraw, ImageFont 
 import requests 
@@ -26,7 +27,7 @@ from retinaface import RetinaFace
 device = "cuda" if torch.cuda.is_available() else "cpu"
 
 # load Gaze-LLE model
-model, transform = torch.hub.load("fkryan/gazelle", "gazelle_dinov2_vitl14_inout")
+model, transform = torch.hub.load("fkryan/gazelle", "gazelle_dinov2_vitl14_inout", use_memory_efficient_attention=False)
 model.eval()
 model.to(device)