@ -172,6 +172,7 @@ image = pipe(
To save VRAM, you can enable CPU offloading
```python
pipe.enable_model_cpu_offload()
pipe.enable_vae_tiling()
```
## Speed Up with LCM-LoRA