Repolode
Submit
Woolverine94

Woolverine94/biniou

A self-hosted web UI that bundles many generative AI modules for images, audio, video, 3D and chat, runnable on CPU-only machines.

Open-source alternative to:Midjourney

View on GitHub
Stars
1,152
Forks
136
Open issues
3
Last commit
2 days ago

What is it?

biniou is a self-hosted web UI for running many generative AI models from one place. It covers text generation through llama-cpp based chatbots and Whisper speech-to-text, image generation and editing with Stable Diffusion, Flux, Kandinsky and PixArt, plus music, audio, video and 3D object generation, all accessed through a Gradio interface in the browser.

The project is built to run on modest hardware. It is set up to work CPU-only with as little as 8GB RAM, with optional CUDA and experimental ROCm acceleration on GNU/Linux. Models are downloaded on first use of each module and stored locally, so the tool can work offline after setup. Installers are provided for several Linux distributions, Windows 10/11, and a Dockerfile for cloud instances.

The web UI includes a control panel for updates, authentication and network sharing, and outputs can be passed from one module to another. Generation settings are saved as metadata in each output file. The project is licensed under GPL-3.0, and each downloaded model carries its own license.

Key features

  • 01Chatbot, image captioning, speech-to-text and translation modules for text tasks
  • 02Image generation and editing with Stable Diffusion, Flux, Kandinsky, ControlNet, inpainting and upscaling
  • 03Music, audio, video and 3D object generation modules
  • 04Runs CPU-only by default, with optional CUDA and experimental ROCm support
  • 05Built-in model management, LoRA and standalone .safetensors or .gguf model support
  • 06Installers for major Linux distributions and Windows, plus a Dockerfile

Use cases

  • Running a local AI studio for image and audio generation on a personal computer without a dedicated GPU
  • Hosting a private generative AI workspace on a cloud instance via Docker
  • Experimenting with several open models in one interface without writing code
  • Offline chatbot and speech-to-text use after models have been downloaded

Quick start

  1. 1

    Pick your operating system in the README installation section and follow its install steps

Things to know

  • Licensed under GPL-3.0, a copyleft license that requires derivative works to be distributed under the same terms.
  • Several modules require 16GB+ RAM, and the macOS install is labeled experimental.
  • Models are downloaded on first use, so an initial setup needs disk space and network access.

In collections

Self-hosted appsRun LLMs locallyAI image generators & Stable Diffusion tools

Topics

#generative-ai#stable-diffusion#text-to-image#llm#self-hosted#gradio#local-ai#image-generation#animatediff#audiocraft

FAQ

What is biniou?
A self-hosted web UI that bundles many generative AI modules for images, audio, video, 3D and chat, runnable on CPU-only machines. biniou is a self-hosted web UI for running many generative AI models from one place. It covers text generation through llama-cpp based chatbots and Whisper speech-to-text, image generation and editing with Stable Diffusion, Flux, Kandinsky and PixArt, plus music, audio, video and 3D object generation, all accessed through a Gradio interface in the browser.
Is biniou free to use? What license does it have?
biniou is open source under the GPL-3.0 license. Check the license terms before using it in commercial products.
Is biniou still maintained?
Actively maintained. The last commit to biniou was on Nov 5, 2026.
What can I use biniou for?
Running a local AI studio for image and audio generation on a personal computer without a dedicated GPU Hosting a private generative AI workspace on a cloud instance via Docker Experimenting with several open models in one interface without writing code Offline chatbot and speech-to-text use after models have been downloaded
How do I install biniou?
Pick your operating system in the README installation section and follow its install steps
Where can I get biniou?
The source code and installation instructions are on GitHub: https://github.com/Woolverine94/biniou

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects