app.py

import gradio as gr
import numpy as np
import random
import spaces
import torch
from diffusers import DiffusionPipeline

dtype = torch.bfloat16
device = "cuda" if torch.cuda.is_available() else "cpu"

# Initialize the model
pipe = DiffusionPipeline.from_pretrained(
    "black-forest-labs/FLUX.1-schnell",
    torch_dtype=dtype
).to(device)

MAX_SEED = np.iinfo(np.int32).max
MAX_IMAGE_SIZE = 2048

# Pattern-specific prompt engineering
def enhance_prompt_for_pattern(prompt):
    """Add specific terms to ensure seamless, tileable patterns."""
    pattern_terms = [
        "seamless pattern",
        "tileable textile design",
        "repeating pattern",
        "high-quality fabric design",
        "continuous pattern",
    ]
    enhanced_prompt = f"{prompt}, {random.choice(pattern_terms)}, suitable for textile printing, high-quality fabric design, seamless edges"
    return enhanced_prompt

@spaces.GPU()
def infer(prompt, seed=42, randomize_seed=False, width=1024, height=1024, 
          num_inference_steps=4, progress=gr.Progress(track_tqdm=True)):
    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    
    # Enhance the prompt for pattern generation
    enhanced_prompt = enhance_prompt_for_pattern(prompt)
    
    generator = torch.Generator().manual_seed(seed)
    image = pipe(
        prompt=enhanced_prompt,
        width=width,
        height=height,
        num_inference_steps=num_inference_steps,
        generator=generator,
        guidance_scale=0.0
    ).images[0]
    
    return image, seed

# Example prompts specifically for pattern generation
examples = [
    "geometric Art Deco shapes in gold and navy",
    "delicate floral motifs with small roses and leaves",
    "abstract watercolor spots in pastel colors",
    "traditional paisley design in earth tones",
    "modern minimalist lines and circles",
]

css = """
#col-container {
    margin: 0 auto;
    max-width: 520px;
}
"""

with gr.Blocks(css=css) as demo:
    with gr.Column(elem_id="col-container"):
        gr.Markdown("""
        # Deradh's AI Pattern Master
        ### Create seamless, tileable patterns for high-quality textile designs
        
        This tool specializes in generating patterns that can be used for fabric printing and textile design.
        Each pattern is optimized to be seamless and repeatable.
        """)
        
        with gr.Row():
            prompt = gr.Text(
                label="Pattern Description",
                show_label=False,
                max_lines=1,
                placeholder="Describe your desired pattern (e.g., 'geometric Art Deco shapes in gold and navy')",
                container=False,
            )
            run_button = gr.Button("Generate Pattern", scale=0)
        
        result = gr.Image(label="Generated Pattern", show_label=True)
        
        with gr.Accordion("Advanced Settings", open=False):
            seed = gr.Slider(
                label="Seed",
                minimum=0,
                maximum=MAX_SEED,
                step=1,
                value=0,
            )
            randomize_seed = gr.Checkbox(label="Randomize seed", value=True)
            
            with gr.Row():
                width = gr.Slider(
                    label="Width",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,
                )
                height = gr.Slider(
                    label="Height",
                    minimum=256,
                    maximum=MAX_IMAGE_SIZE,
                    step=32,
                    value=1024,
                )
            
            with gr.Row():
                num_inference_steps = gr.Slider(
                    label="Number of inference steps",
                    minimum=1,
                    maximum=50,
                    step=1,
                    value=4,
                )
        
        gr.Examples(
            examples=examples,
            fn=infer,
            inputs=[prompt],
            outputs=[result, seed],
            cache_examples="lazy"
        )
        
        gr.on(
            triggers=[run_button.click, prompt.submit],
            fn=infer,
            inputs=[prompt, seed, randomize_seed, width, height, num_inference_steps],
            outputs=[result, seed]
        )

demo.launch()