Update app.py

app.py

@@ -21,7 +21,7 @@ from optimum.quanto import freeze, qfloat8, quantize
 #controlnet = FluxControlNetModel.from_pretrained("alimama-creative/FLUX.1-dev-Controlnet-Inpainting-Alpha", torch_dtype=torch.bfloat16)
 #quantize(controlnet, weights=qfloat8)
 #freeze(controlnet)
-
+controlnet = FluxControlNetModel.from_pretrained("alimama-creative/FLUX.1-dev-Controlnet-Inpainting-Alpha", torch_dtype=torch.bfloat16)
 transformer = FluxTransformer2DModel.from_pretrained(
     "black-forest-labs/FLUX.1-dev", subfolder='transformer', torch_dtype=torch.bfloat16
 )
@@ -34,6 +34,7 @@ pipe = FluxControlNetInpaintingPipeline.from_pretrained(
     "black-forest-labs/FLUX.1-dev",
     text_encoder_2=None,
     transformer=transformer,
+    controlnet=controlnet,
     torch_dtype=torch.bfloat16
 )
 pipe.text_encoder_2 = text_encoder_2
@@ -42,6 +43,8 @@ repo_name = "ByteDance/Hyper-SD"
 ckpt_name = "Hyper-FLUX.1-dev-8steps-lora.safetensors"
 pipe.load_lora_weights(hf_hub_download(repo_name, ckpt_name))
 pipe.fuse_lora(lora_scale=0.125)
+pipe.transformer.to(torch.bfloat16)
+pipe.controlnet.to(torch.bfloat16)
 pipe.to("cuda")
 def can_expand(source_width, source_height, target_width, target_height, alignment):
     if alignment in ("Left", "Right") and source_width >= target_width: