import gradio as gr import torch from diffusers import StableDiffusion3ControlNetPipeline, SD3ControlNetModel, UniPCMultistepScheduler from huggingface_hub import login import os import spaces # Log in to Hugging Face with your token token = os.getenv("HF_TOKEN") login(token=token) # Model IDs for Stable Diffusion 1.5 and ControlNet model_id = "stabilityai/stable-diffusion-3.5-large-turbo" controlnet_id = "InstantX/SD3-Controlnet-Tile" # Load the ControlNet model and Stable Diffusion pipeline controlnet = SD3ControlNetModel.from_pretrained(controlnet_id, torch_dtype=torch.float16) pipe = StableDiffusion3ControlNetPipeline.from_pretrained( model_id, controlnet=controlnet, torch_dtype=torch.float16 ) pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config) pipe = pipe.to("cuda") @spaces.GPU def generate_image(prompt, reference_image,controlnet_conditioning_scale): # Prepare the reference image for ControlNet reference_image = reference_image.convert("RGB").resize((1024, 1024)) # Generate the image with ControlNet conditioning generated_image = pipe( prompt=prompt, control_image=reference_image, controlnet_conditioning_scale=controlnet_conditioning_scale, guidance_scale=7.5, num_inference_steps=50 ).images[0] return generated_image # Set up Gradio interface interface = gr.Interface( fn=generate_image, inputs=[ gr.Textbox(label="Prompt"), gr.Image(type="pil", label="Reference Image (Style)"), gr.Slider(label="Control Net Conditioning Scale",minimum=0,maximum=1), ], outputs="image", title="Image Generation with Stable Diffusion 3.5 and ControlNet", description="Generates an image based on a text prompt and a reference image using Stable Diffusion 3.5 with ControlNet." ) # Launch the Gradio interface interface.launch()