Frequently Asked Questions

General Questions

What is Fooocus?

Fooocus is an image generation tool based on Stable Diffusion XL (SDXL). It provides a simplified interface focused on prompting and generating high-quality images without the complexity of traditional Stable Diffusion interfaces.

Is Fooocus free?

Yes! Fooocus is completely free and open source under the GPL-3.0 license. There are no costs, subscriptions, or limitations.

What makes Fooocus different from other Stable Diffusion interfaces?

Fooocus focuses on simplicity and ease of use. While other interfaces expose hundreds of settings, Fooocus provides a streamlined experience that focuses on what matters most: creating great images. It includes automatic optimizations and quality improvements that work behind the scenes.

Do I need to know how to code to use Fooocus?

No! Fooocus is designed to be user-friendly. The Windows package includes everything you need and can be launched with a simple double-click. No coding knowledge required.

Installation & Setup

What are the system requirements?

Minimum requirements: Windows 10/11, macOS 10.15+, or Linux; Python 3.10+ (included in Windows package); 8GB RAM; 10GB storage. For best performance: 16GB+ RAM, NVIDIA GPU with 8GB+ VRAM. See our Getting Started guide for details.

Do I need to install Python separately?

Windows users: No, the Windows package includes embedded Python. macOS and Linux users: Yes, you need Python 3.10+ installed separately.

How do I install Fooocus on Windows?

Download the Windows ZIP package, extract it to your desired location, and double-click run.bat. That's it! See our Getting Started guide for detailed instructions.

How do I install Fooocus on macOS/Linux?

Install Python 3.10+, then download Fooocus source code, install dependencies with pip install -r requirements.txt, and run python entry_with_update.py. See Getting Started for platform-specific instructions.

Do I need a GPU?

No, but it's highly recommended. Fooocus can run on CPU, but GPU acceleration significantly improves performance. For SDXL models, an NVIDIA GPU with 8GB+ VRAM is ideal.

Models & Downloads

Do I need to download models manually?

No! Fooocus automatically downloads required models on first launch. The SDXL base model will be downloaded automatically when you first run Fooocus.

How large are the models?

The SDXL base model is approximately 7GB. Additional models (refiner, VAE, etc.) may add several more GB. Ensure you have at least 10GB free space, 20GB+ recommended.

Where are models stored?

Models are stored in the models/ directory within your Fooocus installation. You can customize this path in configuration files.

Can I use custom models?

Yes! Place custom SDXL models in the models/checkpoints/ directory. Fooocus will detect and allow you to use them.

Usage & Features

How do I generate an image?

Enter your prompt in the prompt box, optionally adjust settings, and click "Generate". Fooocus will create your image. See our Getting Started guide for a detailed walkthrough.

What are wildcards?

Wildcards allow you to add randomness to prompts. Use __wildcard_name__ syntax, and Fooocus will randomly select from options in wildcard files. Learn more in our Advanced Features guide.

How do I use LoRAs?

Use inline LoRA syntax in your prompt: <lora:lora_name:strength>. Place LoRA files in the models/loras/ directory. See Advanced Features for details.

What is inpainting?

Inpainting allows you to edit specific parts of an image. Load an image, select the area to modify, enter a prompt, and generate. Learn more in Advanced Features.

How do I upscale images?

Fooocus includes upscaling features. Generate your base image, then use the upscaling options to enhance resolution. See Advanced Features for upscaling techniques.

Troubleshooting

Model download fails. What should I do?

Check your internet connection, ensure sufficient disk space, try using the --hf-mirror flag for alternative download source, or manually download models and place them in the correct directory.

I'm getting "Out of Memory" errors. How do I fix this?

Use the --always-low-vram flag, reduce image resolution, close other applications, or use CPU mode if GPU memory is insufficient. See our documentation for memory optimization tips.

Generation is very slow. How can I speed it up?

Ensure GPU acceleration is enabled, check CUDA installation, use appropriate performance presets, and optimize model loading. See documentation for performance tuning.

Fooocus won't start. What's wrong?

Check that Python is installed correctly (for macOS/Linux), verify all dependencies are installed, check system requirements, and review error messages in the console. See our Getting Started guide or community page for help.

Images look blurry or low quality. How do I improve quality?

Use quality modifiers in prompts ("8k", "high detail", "sharp focus"), enable refiner, increase steps, use appropriate CFG scale, and ensure you're using SDXL models. See Features for quality tips.

How do I update Fooocus?

Fooocus includes automatic update checking. To update manually: Windows - download latest release and replace installation; macOS/Linux - use git pull if installed via Git, then pip install -r requirements.txt. Check GitHub releases for latest version.

Configuration & Customization

How do I customize Fooocus settings?

Edit the config.txt file to customize default settings, model paths, and behavior. See our Customization guide for detailed options.

Can I change the default prompt?

Yes! Set default positive and negative prompts in your configuration file. See Customization for details.

How do I create custom styles?

Define custom styles in configuration files or create style files in the styles/ directory. See Customization for style creation guide.

Can I change the output directory?

Yes, configure the output path in config.txt using the output_path setting. See Customization for configuration options.

Advanced Topics

How do I use command-line options?

Run Fooocus with command-line flags like --port 7860, --always-low-vram, --preset anime, etc. See our documentation for complete CLI reference.

What is prompt engineering?

Prompt engineering is the art of crafting effective prompts to get desired results. Use descriptive terms, quality modifiers, style keywords, and negative prompts. Learn techniques in our Advanced Features guide.

How do arrays work?

Arrays allow batch generation by processing multiple values. Use syntax like [[red, green, blue]] flower to generate multiple variations. See Advanced Features for array processing details.

What is the refiner and when should I use it?

The refiner enhances image quality by adding fine details. It's applied after base generation. Enable it in settings or configuration. See Advanced Features for refiner information.

Support & Community

Where can I get help?

Check our documentation, visit the community page, ask on GitHub Discussions, or open an issue on GitHub Issues.

How do I report a bug?

Open an issue on GitHub Issues. Include details about your system, steps to reproduce, and error messages if any.

Can I contribute to Fooocus?

Absolutely! Fooocus is open source and welcomes contributions. Contribute code, documentation, translations, or help test. See our community page for ways to contribute.

Where can I share my creations?

Share on GitHub Discussions, social media with #Fooocus, or check our gallery page for inspiration.

Still Have Questions?

If you couldn't find the answer you're looking for: