Update app.py

app.py

@@ -1,6 +1,6 @@
 import gradio as gr
 import torch
-from diffusers import AutoencoderKL, FluxTransformer2DModel
+from diffusers import AutoencoderKL, FluxTransformer2DModel, FluxFillPipeline
 from diffusers.utils import load_image
 from controlnet_flux import FluxControlNetModel
 from transformer_flux import FluxTransformer2DModel
@@ -12,26 +12,13 @@ import spaces
 from huggingface_hub import hf_hub_download
 from optimum.quanto import freeze, qfloat8, quantize
 
-
-controlnet = FluxControlNetModel.from_pretrained("alimama-creative/FLUX.1-dev-Controlnet-Inpainting-Beta", torch_dtype=torch.bfloat16)
-transformer = FluxTransformer2DModel.from_pretrained(
-    "black-forest-labs/FLUX.1-dev", subfolder='transformer', torch_dtype=torch.bfloat16
-)
-
-pipe = FluxControlNetInpaintingPipeline.from_pretrained(
-    "black-forest-labs/FLUX.1-dev",
-    transformer=transformer,
-    controlnet=controlnet,
+pipe = FluxFillPipeline.from_pretrained(
+    "black-forest-labs/FLUX.1-Fill-dev",
     torch_dtype=torch.bfloat16
-)
-
-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)
+).to("cuda")
+
 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:
         return False
@@ -147,7 +134,7 @@ def inpaint(image, width, height, overlap_percentage, num_inference_steps, resiz
     cnet_image = background.copy()
     cnet_image.paste(0, (0, 0), mask)
 
-    final_prompt = f"{prompt_input} , high quality, 4k"
+    final_prompt = prompt_input
 
     #generator = torch.Generator(device="cuda").manual_seed(42)
 
@@ -155,14 +142,10 @@ def inpaint(image, width, height, overlap_percentage, num_inference_steps, resiz
         prompt=final_prompt,
         height=height,
         width=width,
-        control_image=cnet_image,
-        control_mask=mask,
+        image=cnet_image,
+        mask_image=mask,
         num_inference_steps=num_inference_steps,
-        #generator=generator,
-        controlnet_conditioning_scale=0.9,
-        guidance_scale=3.5,
-        negative_prompt="",
-        true_guidance_scale=3.5,
+        guidance_scale=30,
     ).images[0]
    
     result = result.convert("RGBA")