torch.cuda.empty_cache()

app.py

@@ -396,6 +396,7 @@ def restore_on_gpu(
     x_samples = (einops.rearrange(samples, 'b c h w -> b h w c') * 127.5 + 127.5).cpu().numpy().round().clip(
         0, 255).astype(np.uint8)
     results = [x_samples[i] for i in range(num_samples)]
+    torch.cuda.empty_cache()
 
     # All the results have the same size
     input_height, input_width, input_channel = np.array(input_image).shape