Update app.py

app.py

@@ -9,7 +9,7 @@ from huggingface_hub import hf_hub_download
 device = 'cuda' #if torch.cuda.is_available() else 'cpu'
 torch.cuda.max_memory_allocated(device=device)
 torch.cuda.empty_cache()
-pipe = StableCascadeCombinedPipeline.from_pretrained("stabilityai/stable-cascade", torch_dtype=torch.float16)
+pipe = StableCascadeCombinedPipeline.from_pretrained("stabilityai/stable-cascade", torch_dtype=torch.bfloat16)
 pipe.enable_xformers_memory_efficient_attention()
 pipe = pipe.to(device)
 torch.cuda.empty_cache()