[Tutorial] Introducing Fooocus: Simplified Image Generation for Everyone

fooocus

Fooocus, an innovative image generating software built upon Gradio, brings a fresh approach to Stable Diffusion and Midjourney's designs. It merges the best of both worlds, offering an offline, open-source, and free platform while eliminating the need for manual tweaking, allowing users to focus solely on prompts and images.

One of Fooocus's standout features is its streamlined user experience. It incorporates numerous inner optimizations and quality improvements, relieving users of the burden of complex technical parameters. Instead, it fosters seamless interaction between humans and computers, making image generation an enjoyable endeavor.

Installation of Fooocus has been simplified significantly. With just a few clicks between downloading the software and generating the first image, users can dive right into the creative process. The software demands minimal GPU memory (4GB Nvidia) and is readily available for Windows users.

How to Get Started with Fooocus

Downloading Fooocus

Windows users can easily download Fooocus by clicking here, then following these simple steps:

  • Download the file and unzip it.
  • Run the "run.bat" file.
    Fooocus

Upon the first launch, Fooocus will automatically download default models to the specified folders, catering to various presets. For users employing inpainting features, the software will download its own inpaint control model.

Enhancing Installation Efficiency

For those who already possess the necessary files, accelerating the installation process is feasible by copying them to the designated locations within the software.

Troubleshooting

If encountering errors such as "MetadataIncompleteBuffer" or "PytorchStreamReader" it's likely that the model files are corrupted. In such cases, redownloading the models is recommended.

Performance Insights

On a 16GB RAM, 6GB VRAM (Nvidia 3060 laptop), Fooocus operates at approximately 1.35 seconds per iteration, showcasing impressive speed, especially considering the affordability of laptops equipped with a 3060 GPU.

fooocus
Additionally, it's worth noting that some reports indicate Nvidia driver versions above 532 may cause a tenfold reduction in performance compared to version 531. Users experiencing extended generation times might benefit from reverting to Nvidia Driver 531 Laptop or Nvidia Driver 531 Desktop.

System Requirements

Fooocus mandates a minimum of 4GB Nvidia GPU memory and 8GB system memory. In most cases, Microsoft’s Virtual Swap technique is automatically enabled in Windows installations, eliminating the need for manual configurations.

Colab Setup (Last tested - 2023 Dec 12)

In Colab, modify the last line to execute Fooocus Default, Anime, or Realistic Edition:

!python entry_with_update.py --share or !python entry_with_update.py --preset anime --share or !python entry_with_update.py --preset realistic --share

Please note, the Colab version disables the refiner by default due to resource limitations but guarantees basic text-to-image functionality.

Linux (Anaconda/Miniconda)

For Anaconda/Miniconda users:

git clone https://github.com/lllyasviel/Fooocus.git

cd Fooocus

conda env create -f environment.yaml

conda activate fooocus

pip install -r requirements_versions.txt

Download default models to "Fooocus\models\checkpoints" or let Fooocus automatically handle the downloads using the launcher:

conda activate fooocus

python entry_with_update.py

Or, if you want to open a remote port, use:

conda activate fooocus

python entry_with_update.py --listen

Use python entry_with_update.py --preset anime or python entry_with_update.py --preset realistic for Fooocus Anime/Realistic Edition.

Linux (Python Venv)

Your Linux needs to have Python 3.10 installed, and let's say your Python can be called with the command python3 with your venv system working; you can:

git clone https://github.com/lllyasviel/Fooocus.git

cd Fooocus

python3 -m venv fooocus_env

source fooocus_env/bin/activate

pip install -r requirements_versions.txt

See the above sections for model downloads. You can launch the software with:

source fooocus_env/bin/activate

python entry_with_update.py

Or, if you want to open a remote port, use

source fooocus_env/bin/activate

python entry_with_update.py --listen

Use python entry_with_update.py --preset anime or python entry_with_update.py --preset realistic for Fooocus Anime/Realistic Edition.

Linux (Native System Python)

git clone https://github.com/lllyasviel/Fooocus.git

cd Fooocus

pip3 install -r requirements_versions.txt

See the above sections for model downloads. You can launch the software with:

python3 entry_with_update.py

Or, if you want to open a remote port, use

python3 entry_with_update.py --listen

Use python entry_with_update.py --preset anime or python entry_with_update.py --preset realistic for Fooocus Anime/Realistic Edition.

Linux (AMD GPUs)

Requirements vary for different platforms. Modify the torch package for AMD:

Same with the above instructions. You need to change torch to the AMD version

pip uninstall torch torchvision torchaudio torchtext functorch xformers 

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm5.6

AMD is not intensively tested, however. The AMD support is in beta.

Use python entry_with_update.py --preset anime or python entry_with_update.py --preset realistic for Fooocus Anime/Realistic Edition.

Windows (AMD GPUs)

Windows users with AMD GPUs can follow similar steps as Windows users with other configurations but adapt the content of the run.bat file accordingly.

.\python_embeded\python.exe -m pip uninstall torch torchvision torchaudio torchtext functorch xformers -y

.\python_embeded\python.exe -m pip install torch-directml

.\python_embeded\python.exe -s Fooocus\entry_with_update.py --directml

pause

Then run the run.bat.

AMD is not intensively tested, however. The AMD support is in beta.

For AMD, use .\python_embeded\python.exe entry_with_update.py --directml --preset anime or .\python_embeded\python.exe entry_with_update.py --directml --preset realistic for Fooocus Anime/Realistic Edition.

Mac Setup

For Mac users (M1 or M2), here's an unofficial guideline:

  1. Install the conda package manager and pytorch nightly. Read the Accelerated PyTorch training on Mac Apple Developer guide for instructions. Make sure pytorch recognizes your MPS device.
  2. Open the macOS Terminal app and clone this repository with git clone https://github.com/lllyasviel/Fooocus.git.
  3. Change to the new Fooocus directory, cd Fooocus.
  4. Create a new conda environment, conda env create -f environment.yaml.
  5. Activate your new conda environment, conda activate fooocus.
  6. Install the packages required by Fooocus, pip install -r requirements_versions.txt.
  7. Launch Fooocus by running python entry_with_update.py. (Some Mac M2 users may need python entry_with_update.py --disable-offload-from-vram to speed up model loading/unloading.) The first time you run Fooocus, it will automatically download the Stable Diffusion SDXL models and will take a significant amount of time, depending on your internet connection.
Use python entry_with_update.py --preset anime or python entry_with_update.py --preset realistic for Fooocus Anime/Realistic Edition.

Minimal Requirement

fooocus

fooocus


For more details: https://github.com/lllyasviel/Fooocus#readme

Comments

Post a Comment

Leave your comment here...

Popular posts from this blog

[Tutorial] Scenario AI: Create Consistent Characters

Image
  In this short tutorial, I will show you how to use Scenario AI, a new AI for quick and easy generating game assets. Software Scenario (free tier available) Photoshop  or Gimp (or any photo editing software) Magnific AI (optional) Pricing Scenario AI offers 4 pricing plans, including a free option, perfect for this tutorial. How to use Scenario AI Choose a model: Go to the Models tab in Scenario. Select a model that is in the vibe of the character you want to create. Don't worry about finding the perfect one yet, you can blend and remix later. Click "Generate with this model" to get a basic idea of the style. Prompt your character: Use the prompt box to describe your character, including details like age, hair color, clothing type, etc. Add prompts for front, side, and back views to ensure the model generates the character from multiple angles. Use the Prompt Builder or Prompt Spark features for inspiration if needed. Don't forget negative prompts to avoid unwanted e...
Read more

[Tutorial] XeroGen: Generate High-Quality Prompts for Stable Diffusion, Midjourney, Dall-E, and Leonardo AI

Image
What is XeroGen? XeroGen  is a subscription-based platform that combines prompt generators for Stable Diffusion, Midjourney, Dall-E, and Leonardo AI into one. It helps you create high-quality prompts for these AI image generators quickly and easily, saving you time and effort. How does XeroGen work? Specify your desired image: Tell XeroGen what you want to generate, like "a cyberpunk cityscape" or "a portrait of a cat in a hat." Choose your AI image generator: Select the AI image generator you want to use, such as Stable Diffusion or Midjourney. Get prompts: XeroGen will generate several prompts based on your preferences. You can also specify the number of prompts you want (up to five). Fine-tune the prompts (optional): You can customize the prompts to your liking using the available sliders and options. Use the prompts in your AI image generator: Copy the prompts and paste them into your chosen AI image generator to create your artwork! Benefits of using XeroGen: S...
Read more