LlamaIndex
LlamaIndex builds document parsing and extraction software plus open-source tools for AI applications, helping teams turn enterprise documents into usable data for AI agents and workflows.

Fooocus is a free, offline, open-source image generator built on the Stable Diffusion XL architecture, aimed at users who want to focus on prompts and images rather than parameter tuning. It runs through a Gradio interface and is distributed from its official GitHub repository.
Fooocus rethinks the usual image generator workflow. Instead of exposing users to a wall of settings, it handles much of the technical work behind the scenes. The project describes an offline GPT-2 based prompt processing engine and a set of sampling improvements that aim to produce good results whether a prompt is as short as "house in garden" or as long as 1,000 words. That approach puts Fooocus in the same conversational territory as hosted services such as Midjourney, but without the online dependency.
Setup is deliberately short. According to the project, the number of mouse clicks between pressing download and generating a first image is strictly limited to fewer than three. The minimum GPU memory requirement is 4GB on Nvidia hardware, which keeps the barrier to entry relatively low for a local SDXL-based tool. Users who run into a "RuntimeError: CPUAllocator" message are advised to keep at least 40GB of free space on each drive.
Fooocus covers the core operations users expect from an image generator. From an input image, users can upscale, create subtle or strong variations, and inpaint or outpaint in several directions. The project notes that it uses its own inpaint algorithm and models rather than the standard SDXL inpaint method. Image Prompt is handled through a dedicated algorithm as well. Style and guidance controls live under the Advanced settings, and prompts can include inline LoRAs written as flower <lora:sunflowers:1.2>, provided the LoRA file sits in the models/loras directory.
Fooocus is now in a state of limited long-term support, with future updates focused on bug fixes. The maintainers consider the existing functionality to be nearly free of programming issues, crediting contributor mashb1t for much of that work. There are no current plans to migrate to or incorporate newer model architectures, though the project leaves room for that to change if the open-source community converges on a single dominant method for image generation.
Users who want to work with newer models such as Flux are pointed toward other options, including WebUI Forge, which comes from the same developer, as well as ComfyUI and SwarmUI. Several forks of Fooocus are also available for experimentation, including fenneishi/Fooocus-Control, runew0lf/RuinedFooocus, MoonRide303/Fooocus-MRE, and mashb1t/Fooocus.
The project warns that many fake websites appear in Google results for "fooocus." Domains such as fooocus.com, fooocus.net, fooocus.co, fooocus.ai, fooocus.org, fooocus.pro, and fooocus.one are described as fake and unrelated to the project. The GitHub repository is the only official source, and Fooocus is presented as 100% non-commercial offline open-source software.
Fooocus supports interface translation through JSON files placed in the language folder. A --language argument tells the software which file to load, and the project invites users to contribute new translation files for languages that do not yet exist. That localization path gives non-English speakers a way to adapt the interface without modifying the core code.
Fooocus is described as offline, open source, and free, and the project states it is 100% non-commercial offline open-source software. No paid tiers or pricing are listed on the official page.
Features, pricing and availability may change. Always verify details on the official website. Maintained by steven · Last updated Oct 16, 2026 →
LlamaIndex builds document parsing and extraction software plus open-source tools for AI applications, helping teams turn enterprise documents into usable data for AI agents and workflows.
Flowith is an agentic AI workspace that combines a whiteboard-style canvas, access to 40+ image and video models, and mini apps for design, e-commerce, and social content.
Fine-tune open models faster with far less memory
Groq is an AI inference cloud for developers and enterprises that need low-latency model responses at scale, delivered through custom LPU hardware and a tiered platform spanning bare metal, inference, and enterprise controls.