from diffusers import StableDiffusionPipeline, StableDiffusionImg2ImgPipeline, DPMSolverMultistepScheduler import gradio as gr import torch from PIL import Image model_id = 'Dunkindont/Foto-Assisted-Diffusion-FAD_V0/' prefix = '' scheduler = DPMSolverMultistepScheduler.from_pretrained(model_id, subfolder="FAD768-Diffuser Model/scheduler") pipe = StableDiffusionPipeline.from_pretrained( model_id, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, scheduler=scheduler) pipe_i2i = StableDiffusionImg2ImgPipeline.from_pretrained( model_id, torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32, scheduler=scheduler) if torch.cuda.is_available(): pipe = pipe.to("cuda") pipe_i2i = pipe_i2i.to("cuda") def error_str(error, title="Error"): return f"""#### {title} {error}""" if error else "" def _parse_args(prompt, generator): parser = argparse.ArgumentParser( description="making it work." ) parser.add_argument( "--no-half-vae", help="no half vae" ) cmdline_args = parser.parse_args() command = cmdline_args.command conf_file = cmdline_args.conf_file conf_args = Arguments(conf_file) opt = conf_args.readArguments() if cmdline_args.config_overrides: for config_override in cmdline_args.config_overrides.split(";"): config_override = config_override.strip() if config_override: var_val = config_override.split("=") assert ( len(var_val) == 2 ), f"Config override '{var_val}' does not have the form 'VAR=val'" conf_args.add_opt(opt, var_val[0], var_val[1], force_override=True) def inference(prompt, guidance, steps, width=768, height=768, seed=0, img=None, strength=0.5, neg_prompt="", auto_prefix=False): generator = torch.Generator('cuda').manual_seed(seed) if seed != 0 else None prompt = f"{prefix} {prompt}" if auto_prefix else prompt try: if img is not None: return img_to_img(prompt, neg_prompt, img, strength, guidance, steps, width, height, generator), None else: return txt_to_img(prompt, neg_prompt, guidance, steps, width, height, generator), None except Exception as e: return None, error_str(e) def txt_to_img(prompt, neg_prompt, guidance, steps, width, height, generator): result = pipe( prompt, negative_prompt = neg_prompt, num_inference_steps = int(steps), guidance_scale = guidance, width = width, height = height, generator = generator) return result.images[0] def img_to_img(prompt, neg_prompt, img, strength, guidance, steps, width, height, generator): ratio = min(height / img.height, width / img.width) img = img.resize((int(img.width * ratio), int(img.height * ratio)), Image.LANCZOS) result = pipe_i2i( prompt, negative_prompt = neg_prompt, init_image = img, num_inference_steps = int(steps), strength = strength, guidance_scale = guidance, width = width, height = height, generator = generator) return result.images[0] def fake_safety_checker(images, **kwargs): return result.images[0], [False] * len(images) pipe.safety_checker = fake_safety_checker css = """.main-div div{display:inline-flex;align-items:center;gap:.8rem;font-size:1.75rem}.main-div div h1{font-weight:900;margin-bottom:7px}.main-div p{margin-bottom:10px;font-size:94%}a{text-decoration:underline}.tabs{margin-top:0;margin-bottom:0}#gallery{min-height:20rem} """ with gr.Blocks(css=css) as demo: gr.HTML( f"""

Foto Assisted Diffusion (FAD)

Demo for (FAD) Stable Diffusion model by Dunkindont. {"" if prefix else ""} Running on {"GPU 🔥" if torch.cuda.is_available() else f"CPU 🔥"}.

• Model is meant to mimic a modern HDR photography style. • Trained on 600 HDR images on SD1.5 • Merged with one of my own models for illustrations and drawings for miniscule uses outside of photography • No hi-res fix required, can generate natively from supported resolutions (See Supported Resolutions Tab)
Please use this prompt template below to get an example of the desired generation results:
Important note: You can use the model at 512x512 but the results will likely be undesirable.
Prompt: medium portrait (close up:0.5) of a Beautiful scottish ginger woman, (messy bun), sitting on a wooden chair, wearing a fancy Victorian era dress, seductively posing, pale skin, dark red lips, dark eye shadow, ornate pendant, sitting in front of a large stone fireplace, candle burning, in a luxurious fantasy castle, side lighting, cinematic, Renaissance style, (Fujifilm XT3:1.1), (high detailed face:1.3), perfect hands Negative Prompt: Negative prompt: (deformed mouth), (deformed lips), (deformed eyes), (cross-eyed), (deformed iris), (deformed hands), lowers, 3d render, cartoon, long body, wide hips, narrow waist, disfigured, ugly, cross eyed, squinting, grain, Deformed, blurry, bad anatomy, poorly drawn face, mutation, mutated, extra limb, ugly, (poorly drawn hands), missing limb, floating limbs, disconnected limbs, malformed hands, blur, out of focus, long neck, disgusting, poorly drawn, mutilated, , mangled, old, surreal, ((text))

Have Fun & Enjoy Website
""" ) with gr.Row(): with gr.Column(scale=55): with gr.Group(): with gr.Row(): prompt = gr.Textbox(label="Positive Prompt", show_label=False, max_lines=2, placeholder=f"{prefix} [your prompt]").style(container=False) generate = gr.Button(value="Generate").style(rounded=(False, True, True, False)) image_out = gr.Image(height=512) error_output = gr.Markdown() with gr.Column(scale=45): with gr.Tab("Options"): with gr.Group(): neg_prompt = gr.Textbox(label="Negative prompt", placeholder="What to exclude from the image") auto_prefix = gr.Checkbox(label="Prefix styling tokens automatically (RAW photo,)", value=prefix, visible=prefix) with gr.Row(): guidance = gr.Slider(label="Guidance scale", value=7.5, maximum=15) steps = gr.Slider(label="Steps", value=25, minimum=2, maximum=75, step=1) with gr.Row(): width = gr.Slider(label="Width", value=512, minimum=64, maximum=1024, step=8) height = gr.Slider(label="Height", value=512, minimum=64, maximum=1024, step=8) seed = gr.Slider(0, 2147483647, label='Seed (0 = random)', value=0, step=1) with gr.Tab("Image to image"): with gr.Group(): image = gr.Image(label="Image", height=256, tool="editor", type="pil") strength = gr.Slider(label="Transformation strength", minimum=0, maximum=1, step=0.01, value=0.5) auto_prefix.change(lambda x: gr.update(placeholder=f"{prefix} [your prompt]" if x else "[Your prompt]"), inputs=auto_prefix, outputs=prompt, queue=False) inputs = [prompt, guidance, steps, width, height, seed, image, strength, neg_prompt, auto_prefix] outputs = [image_out, error_output] prompt.submit(inference, inputs=inputs, outputs=outputs) generate.click(inference, inputs=inputs, outputs=outputs) demo.queue(concurrency_count=1) demo.launch()