enable CPU offloading

app.py

@@ -1,6 +1,6 @@
 import torch
 import spaces
-from diffusers import ControlNetUnionModel, AutoencoderKL, UNet2DConditionModel
+from diffusers import ControlNetUnionModel, AutoencoderKL
 import gradio as gr
 
 from pipeline.mod_controlnet_tile_sr_sdxl import StableDiffusionXLControlNetTileSRPipeline
@@ -10,7 +10,6 @@ from pipeline.util import (
     calculate_overlap,
     create_hdr_effect,
     progressive_upscale,
-    quantize_8bit,
     select_scheduler,
 )
 
@@ -27,11 +26,7 @@ pipe = StableDiffusionXLControlNetTileSRPipeline.from_pretrained(
     model_id, controlnet=controlnet, vae=vae, torch_dtype=torch.float16, use_safetensors=True, variant="fp16"
 ).to(device)
 
-# unet = UNet2DConditionModel.from_pretrained(model_id, subfolder="unet", variant="fp16", use_safetensors=True)
-# quantize_8bit(unet)  # << Enable this if you have limited VRAM
-# pipe.unet = unet
-
-#pipe.enable_model_cpu_offload()  # << Enable this if you have limited VRAM
+pipe.enable_model_cpu_offload()  # << Enable this if you have limited VRAM
 pipe.enable_vae_tiling() # << Enable this if you have limited VRAM
 pipe.enable_vae_slicing() # << Enable this if you have limited VRAM