app.py

import gradio as gr
import requests
import base64
from PIL import Image
import io
import json

# Mock function to simulate image generation
def generate_image(prompt, negative_prompt, guidance_scale, num_inference_steps, width, height, seed):
    """
    Generate an image using NewBie-AI/NewBie-image-Exp0.1
    In a real implementation, this would connect to the actual model API
    """
    try:
        # This is a placeholder for actual model inference
        # In production, you would replace this with actual API calls to NewBie-AI
    
    # For demonstration, we'll create a simple gradient image
    # In reality, this would call: https://huggingface.co/NewBie-AI/NewBie-image-Exp0.1
        
        # Create a colorful gradient image
        import numpy as np
        img = np.zeros((height, width, 3), dtype=np.uint8)
        
        # Create gradient
        for i in range(height):
            for j in range(width):
                img[i, j, 0] = int(255 * i / height)
        img[i, j, 1] = int(255 * j / width)
        img[i, j, 2] = int(255 * (i + j) / (height + width))
        
        return img
    except Exception as e:
        raise gr.Error(f"Image generation failed: {str(e)}")

def generate_images_interface(
    prompt,
    negative_prompt="",
    guidance_scale=7.5,
    num_inference_steps=20,
    width=512,
    height=512,
    seed=-1,
    batch_size=1
):
    """
    Interface function for image generation
    """
    # Validate inputs
    if not prompt or len(prompt.strip()) == 0:
        raise gr.Warning("Please enter a prompt to generate an image")
    
    if width < 64 or width > 1024:
        raise gr.Warning("Width must be between 64 and 1024")
    if height < 64 or height > 1024:
        raise gr.Warning("Height must be between 64 and 1024")
    
    generated_images = []
    for i in range(batch_size):
        image = generate_image(
            prompt=prompt,
            negative_prompt=negative_prompt,
            guidance_scale=guidance_scale,
            num_inference_steps=num_inference_steps,
            width=width,
            height=height,
            seed=seed if seed != -1 else None
        )
        generated_images.append(image)
    
    if batch_size == 1:
        return generated_images[0]
    else:
        return generated_images

def main():
    with gr.Blocks() as demo:
        gr.Markdown("# 🎨 NewBie-AI Image Generator")
        gr.Markdown("Create stunning AI-generated images with full customization")
        
        with gr.Row():
            with gr.Column(scale=1):
                with gr.Group():
                    gr.Markdown("### 📝 Text Prompts")
                    prompt_input = gr.Textbox(
                        label="Prompt",
                        placeholder="Describe the image you want to create...",
                        lines=3,
                        value="A beautiful landscape with mountains and a lake"
                    )
                    negative_prompt_input = gr.Textbox(
                        label="Negative Prompt",
                        placeholder="What you don't want in the image...",
                    )
                    
                    with gr.Row():
                        width_slider = gr.Slider(
                            minimum=64,
                            maximum=1024,
                            value=512,
                            step=64,
                            label="Width"
                    )
                    height_slider = gr.Slider(
                            minimum=64,
                            maximum=1024,
                            value=512,
                            label="Height"
                    )
                    
                    with gr.Row():
                        guidance_scale_slider = gr.Slider(
                            minimum=1.0,
                            maximum=20.0,
                            value=7.5,
                            step=0.5,
                            label="Guidance Scale"
                    )
                    
                    with gr.Row():
                        inference_steps_slider = gr.Slider(
                            minimum=1,
                            maximum=50,
                            value=20,
                            step=1,
                            label="Inference Steps"
                    )
                    
                    with gr.Row():
                        seed_input = gr.Number(
                            value=-1,
                            label="Seed (-1 for random)"
                    )
                    
                    with gr.Row():
                        batch_size_dropdown = gr.Dropdown(
                            choices=["1", "2", "4"],
                            value="1",
                            label="Batch Size"
                    )
                
                with gr.Group():
                    gr.Markdown("### ⚙️ Advanced Settings")
                        num_inference_steps = gr.Number(
                            value=20,
                            label="Number of Inference Steps"
                    )
                
                generate_btn = gr.Button(
                    "Generate Image 🎨",
                    variant="primary",
                    size="lg"
                )
            
            with gr.Column(scale=1):
                with gr.Group():
                    gr.Markdown("### 🖼️ Generated Images")
                    output_gallery = gr.Gallery(
                        label="Generated Images",
                        columns=2,
                        height=500
                    )
        
        # Event handling with Gradio 6 syntax
        generate_btn.click(
            fn=generate_images_interface,
            inputs=[
                prompt_input,
                negative_prompt_input,
                guidance_scale_slider,
                        inference_steps_slider,
                        width_slider,
                        height_slider,
                        seed_input
                    ],
            outputs=[output_gallery],
            api_visibility="public"
        )
        
        # Examples section
        gr.Examples(
            examples=[
                [
                    "A majestic dragon flying over a medieval castle at sunset",
                    "blurry, low quality",
                    7.5,
                    20,
                    512,
                    512,
                    -1
                ],
                [
                    "A futuristic cityscape with flying cars and neon lights",
                    "watermark, signature",
                    8.0,
                    25,
                    768,
                    768,
                    42
                ],
                [
                    "An astronaut riding a horse on Mars, photorealistic",
                    "cartoon, animated",
                    12.0,
                    30,
                    1024,
                    1024,
                    123
                ]
            ],
            inputs=[
                prompt_input,
                negative_prompt_input,
                guidance_scale_slider,
                        inference_steps_slider,
                        width_slider,
                        height_slider,
                        seed_input
            ],
            outputs=[output_gallery],
            fn=generate_images_interface,
            cache_examples=True
        )
        
        gr.Markdown("---")
        gr.HTML('<div style="text-align: center; padding: 20px; font-size: 14px; color: #666;">Built with <a href="https://huggingface.co/spaces/akhaliq/anycoder" style="color: #666; text-decoration: none;">anycoder</a></div>')
    
    # Launch with Gradio 6 syntax - ALL parameters go here
    demo.launch(
        theme=gr.themes.Soft(
            primary_hue="indigo",
            secondary_hue="blue",
            neutral_hue="slate",
            font=gr.themes.GoogleFont("Inter"),
            text_size="lg",
            spacing_size="lg",
            radius_size="md"
        ).set(
            button_primary_background_fill="*primary_600",
            button_primary_background_fill_hover="*primary_700",
            block_title_text_weight="600",
        ),
        css="""
        .gradio-container {
            max-width: 1200px !important;
        }
        .darktest {
            background-color: #f8fafc;
            padding: 20px;
            border-radius: 8px;
        }
        .cool-col {
            border: 1px solid #e2e8f0;
            border-radius: 8px;
            padding: 20px;
        }
        """,
        footer_links=[{"label": "Built with anycoder", "url": "https://huggingface.co/spaces/akhaliq/anycoder"}
    )

if __name__ == "__main__":
    main()