README.md

metadata

title: Gokaygokay Flux Seamless Texture LoRA
emoji: ⚡
colorFrom: purple
colorTo: blue
sdk: gradio
sdk_version: 6.1.0
app_file: app.py
pinned: false

Flux Seamless Texture LoRA

Generate seamless textures using the Flux LoRA model. This Space provides a modern interface with advanced controls, presets, batch generation, gallery, and MCP (Model Context Protocol) integration.

Features

• 🎨 Advanced Generation Controls: Fine-tune guidance scale, steps, seed, dimensions, and more
• 🎯 Presets: Pre-configured settings for common texture types (wood, fabric, metal, stone, etc.)
• 📦 Batch Generation: Generate multiple textures at once with progress tracking
• 🖼️ Gallery: View and download all generated textures
• 📜 History: Track and reuse previous generation parameters
• 🔌 MCP Integration: Access via Model Context Protocol for AI tool integration
• 🎨 Modern UI: Clean, responsive interface built with Gradio 6

Quick Start

1. Enter a prompt describing the texture you want
2. Adjust parameters or select a preset
3. Click "Generate" to create your texture
4. View results in the Gallery tab
5. Access history in the History tab

Usage

Basic Generation

1. Go to the Generate tab
2. Enter your prompt (e.g., "seamless wood texture, natural grain")
3. Optionally select a preset from the dropdown
4. Adjust parameters as needed:
   • Guidance Scale: Controls how closely the model follows the prompt (1-20)
   • Inference Steps: Number of denoising steps (10-100)
   • Seed: Random seed for reproducibility (-1 for random)
   • Width/Height: Image dimensions (must be multiples of 8)
5. Click Generate

Using Presets

Select a preset from the dropdown to automatically apply optimized settings:

• Wood Texture
• Fabric Texture
• Metal Texture
• Stone Texture
• Brick Texture
• Leather Texture
• Concrete Texture
• Marble Texture

Batch Generation

1. Go to the Batch Generate tab
2. Enter multiple prompts, one per line
3. Set common parameters
4. Click Generate Batch
5. Monitor progress and view results

Gallery

• View all generated textures in a grid
• Click Refresh Gallery to update
• Download all images as a ZIP file

History

• View generation history with timestamps
• Reuse parameters from previous generations
• Export history data

MCP Integration

This Space supports Model Context Protocol (MCP) for integration with AI tools.

Configuration

See docs/MCP.md for detailed setup instructions.

Quick Setup:

1. Add to your MCP client configuration:

{
  "mcpServers": {
    "flux-texture-lora": {
      "url": "https://huggingface.co/spaces/gokaygokay/Flux-Seamless-Texture-LoRA/gradio_api/mcp/",
      "type": "http"
    }
  }
}

2. Restart your MCP client
3. Use tools like generate_texture, get_presets, etc.

Available MCP Tools

• generate_texture - Generate a texture with parameters
• generate_batch_textures - Generate multiple textures
• get_presets - List available presets
• get_history - Get generation history
• download_image - Get image information

See docs/MCP.md for complete documentation.

Parameters

Generation Parameters

• Prompt: Text description of the desired texture
• Negative Prompt: What to avoid in generation
• Guidance Scale (1-20): How closely to follow the prompt
• Inference Steps (10-100): Number of denoising steps
• Seed (-1 or positive): Random seed for reproducibility
• Width/Height (256-2048, multiple of 8): Image dimensions
• CFG Scale (1-20): Classifier-free guidance scale
• LoRA Strength (0-2): Strength of LoRA adaptation

Installation

Hugging Face Spaces (recomendado)

• Entry point: app.py (já configurado no YAML do topo deste README).
• Secrets: se o modelo for privado, adicione HF_TOKEN em Settings → Secrets do Space.
• Storage: se você habilitar Persistent Storage, os arquivos gerados serão gravados em /data automaticamente (fallback para o diretório do repo se /data não existir).
• Queue: o app roda com queue() habilitado para suportar inferência demorada sem travar o servidor.

Desenvolvimento local (opcional)

1. Clone the repository:

git clone https://huggingface.co/spaces/gokaygokay/Flux-Seamless-Texture-LoRA
cd Flux-Seamless-Texture-LoRA

2. Install dependencies:

pip install -r requirements.txt

3. Set up environment variables (optional):

cp .env.example .env
# Edit .env with your settings

4. Run the application:

python app.py

Project Structure

.
├── app.py                 # Main Gradio application
├── requirements.txt       # Python dependencies
├── config/
│   └── settings.py       # Configuration settings
├── src/
│   ├── model_handler.py  # Model management
│   ├── image_processor.py # Image processing
│   ├── presets.py        # Texture presets
│   ├── utils.py          # Utility functions
│   └── mcp_server.py     # MCP server implementation
├── tests/                # Unit tests
├── docs/                 # Documentation
└── assets/               # Assets and examples

Documentation

• API Documentation - Complete API reference
• MCP Integration - MCP setup and usage
• Gradio 6 Migration - Migration details

Development

Running Tests

pytest tests/

Code Style

• Type hints required
• Google-style docstrings
• Follow PEP 8

Troubleshooting

Common Issues

Generation fails:

• Check that dimensions are multiples of 8
• Verify prompt is not empty
• Ensure parameters are within valid ranges

MCP not working:

• Verify Space URL is correct
• Check that MCP is enabled
• Ensure client configuration is correct

Images not saving:

• Check write permissions
• Verify OUTPUT_DIR exists
• Check disk space

Contributing

Contributions are welcome! Please:

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Add tests
5. Submit a pull request

License

See the repository for license information.

Acknowledgments

• Flux model by Black Forest Labs
• Gradio for the UI framework
• Hugging Face for hosting

Changelog

Version 2.0.0 (Current)

• ✅ Migrated to Gradio 6
• ✅ Added MCP integration
• ✅ Modernized UI with tabs
• ✅ Added batch generation
• ✅ Added gallery and history
• ✅ Added presets system
• ✅ Comprehensive documentation

Version 1.0.0

• Initial release with basic generation

---

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference