LIVE
Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|Failed startups' data finds a second life training AI models28/10/26|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|New Benchmark Tests Whether LLMs Can Uncover Hidden Physical Laws27/10/26|Liquid AI unveils LFM3, a base for specialized multimodal models27/10/26 · Liquid AI|US Police Departments Use AI to Flag 'Suspicious' Driving Behavior27/10/26|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|NVIDIA introduces Nemotron 3.5 Transcribe, a promptable multilingual transcription model28/10/26 · NVIDIA|Failed startups' data finds a second life training AI models28/10/26|Anthropic shares early results on using Claude to accelerate drug development27/10/26 · Anthropic|GEPA-Curator: Filtering Failures to Better Train LLM Agents27/10/26|New Benchmark Tests Whether LLMs Can Uncover Hidden Physical Laws27/10/26|Liquid AI unveils LFM3, a base for specialized multimodal models27/10/26 · Liquid AI|US Police Departments Use AI to Flag 'Suspicious' Driving Behavior27/10/26|
AppWatch outofficial publisher

Fooocus

App : A product you use directly

Ultra-simplified local interface to generate high-quality SDXL images without technical adjustments.

The best first step into local and private image generation, as long as you stay within SDXL.

GitHub · 53,682 ★
Overall score
7.7 /10
Security: nothing to flag
Utility
8.0
Reliability
4.0
Security
No caveat on record
Ease
5 min · Comfortable

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Generating images with a local AI tool usually requires configuring dozens of technical parameters, which discourages most non-specialist users. Fooocus solves this problem by reducing the interface to the essentials: a text field to describe the desired image, and a button to start generation.

The software works entirely offline, on your own machine, without sending your data to a remote server. It is built on the Stable Diffusion XL architecture and includes a GPT-2-based prompt processing engine that automatically improves results even with short or vague descriptions. Installation takes fewer than three clicks on Windows, and models are downloaded automatically on first launch.

Fooocus is aimed at people who want to generate quality images locally, without a paid online subscription and without mastering the advanced settings of tools like ComfyUI or Automatic1111. It suits both curious creatives and users who care about the privacy of their work.

However, Fooocus is not suited for users who want to work with newer architectures such as Flux, or who need fine-grained control over every generation parameter. The project is now in maintenance mode: only bug fixes are planned, with no migration to newer models.

What it offers

  • ·Local text-to-image generation, with no internet connection or data sent to a remote server.
  • ·GPT-2-based prompt processing engine that automatically improves output quality regardless of description length.
  • ·Image inpainting and outpainting using a proprietary algorithm distinct from the standard SDXL method.
  • ·Image Prompt feature that uses a reference image as a visual guide, via Fooocus's own algorithm.
  • ·Upscaling and variation of existing images (x1.5 or x2 upscale, subtle or strong variations).
  • ·Three preconfigured launch profiles: default, anime, and realistic, also selectable from the browser interface.
  • ·Compatibility with SDXL models available on Civitai, to diversify generation styles.
  • ·Face swapping on an image (FaceSwap) via InsightFace, accessible from the Image Prompt menu.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
windows, linux, macos
Licence
GPL-3.0
FR interface
No

Security and data

Rating
Security: nothing to flag
Your data
Data stays local
Known incidents
Projet en mode LTS (corrections seulement).
Maintenance
Inactive for over six months
Last release
12 August 2024

Links and repository checked on 25 October 2026 · added on 1 August 2023

How to install Fooocus

1. Download the official archive only from the project's GitHub page (beware of fake websites). 2. Extract the archive to a folder of your choice. 3. Run "run.bat" (Windows); SDXL models are downloaded automatically on first launch. 4. On Linux, clone the repository, create an Anaconda or Python venv environment, then run "python entry_with_update.py". 5. Make sure your Nvidia GPU has at least 4 GB of VRAM and your system has at least 8 GB of RAM. 6. If you encounter "RuntimeError: CPUAllocator" errors, enable Windows virtual memory and free up at least 40 GB of disk space.
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

In the same category

Whole category Photo & image →