airunner

airunner

Stable Diffusion and LLMs offline on your own hardware

Stars: 264

Visit
 screenshot

AI Runner is a multi-modal AI interface that allows users to run open-source large language models and AI image generators on their own hardware. The tool provides features such as voice-based chatbot conversations, text-to-speech, speech-to-text, vision-to-text, text generation with large language models, image generation capabilities, image manipulation tools, utility functions, and more. It aims to provide a stable and user-friendly experience with security updates, a new UI, and a streamlined installation process. The application is designed to run offline on users' hardware without relying on a web server, offering a smooth and responsive user experience.

README:

Banner Discord PyPi GitHub GitHub last commit GitHub issues GitHub closed issues GitHub pull requests GitHub closed pull requests


AI Runner

AI Runner is an interface which allows you to run open-source large language models (LLM) and AI image generators (Stable Diffusion) on your own hardware.

It is designed to be easy to use, with a simple and intuitive interface that allows you to run AI models without the need for a web server or cloud service.

It has been optimized for speed and efficiency, allowing you to generate images and have conversations with chatbots in real-time.

Stable Diffusion

img.png

Customizable Chatbots with Moods and Personalities

img_1.png


⭐ Features

AI Runner is an AI interface which allows you to run open-source large language models (LLM) and AI image generators (Stable Diffusion) on your own hardware.

Feature Description
🗣️ LLMs and communication
Voice-based chatbot conversations Have conversations with a chatbot using your voice
Text-to-speech Convert text to spoken audio
Speech-to-text Convert spoken audio to text
Customizable chatbots with LLMs Generate text using large language models
RAG on local documents and websites Interact with your local documents using an LLM
🎨 Image Generation
Stable Diffusion (all versions) Generate images using Stable Diffusion
Drawing tools Turn sketches into art
Text-to-Image Generate images from textual descriptions
Image-to-Image Generate images based on input images
🖼️ Image Manipulation
Inpaint and Outpaint Modify parts of an image while maintaining context
Controlnet Control image generation with additional input
LoRA Efficiently fine-tune models with LoRA
Textual Embeddings Use textual embeddings for image generation control
Image Filters Blur, film grain, pixel art and more
🔧 Utility
Run offline, locally Run on your own hardware without internet
Fast generation Generate images in ~2 seconds (RTX 2080s)
Run multiple models at once Utilize multiple models simultaneously
Dark mode Comfortable viewing experience in low-light environments
Infinite scrolling canvas Seamlessly scroll through generated images
NSFW filter toggle Help control the visibility of NSFW content
NSFW guardrails toggle Help prevent generation of LLM harmful content
Fully customizable Easily adjust all parameters
Fast load time, responsive interface Enjoy a smooth and responsive user experience
Pure python No reliance on a webserver, pure python implementation

💻 System Requirements

Minimum system requirements

  • OS: Linux
  • Processor: Intel i5 or equivalent
  • Memory: 16 GB RAM
  • Graphics: 2080s RTX or higher
  • Network: Broadband Internet connection required for setup
  • Storage: 130 GB available space

Recommended system specs

  • OS: Linux
  • Processor: Intel i7 or equivalent
  • Memory: 30 GB RAM
  • Graphics: 4090 RTX or higher
  • Network: Broadband Internet connection required for setup
  • Storage: 130 GB available space

🔧 Running AI Runner from source

Running AI Runner from source is recommended for developers and users who want to test the latest features.

Linux

Install prerequisites

sudo apt update
sudo apt install -y fonts-noto-color-emoji
sudo apt install -y libportaudio2
sudo apt install -y libxcb-cursor0
sudo apt install -y espeak
sudo apt install -y xclip
sudo apt install -y git
sudo apt install -y python3-pip
sudo apt install -y python3.10-venv

Clone the repository

git clone https://github.com/Capsize-Games/airunner.git
cd airunner

Create a virtual environment

python3 -m venv airunner
source airunner/bin/activate

Install the required packages

pip install -e .

Run AI Runner

airunner

Using AI Runner as a Python library

AI Runner can be used as a Python library in order to extend its functionality.

pip install airunner

Now you can import AI Runner into your Python scripts.

from airunner import airunner

Packaging AI Runner

Detailed packaging instructions can be found in the wiki.


Models

AI Runner installs all of the models required to run a chatbot with text-to-speech and speech-to-text capabilities, as well as the core models required for Stable Diffusion. However, you must supply your own art generator models.

You can download models from Huggingface.co or civitai.com.

The supported Stable Diffusion models are:

  • SD 1.5
  • SDXL 1.0
  • SDXL Turbo

Models must be placed in their respective directories in the airunner directory.

~/.local/share/airunner
├── art
│   ├── models
│   │   ├── SD 1.5
│   │   │   ├── lora
│   │   │   └── embeddings
│   │   ├── SDXL 1.0
│   │   │   ├── lora
│   │   │   └── embeddings
│   │   └── SDXL Turbo
│   │       ├── lora
│   │       └── embeddings

Privacy and Security

Although AI Runner v3.0 is built with Huggingface libraries, we have taken care to strip the application of any telemetry or tracking features.


Internet access

Only the setup wizard needs access to the internet in order to download the required models.

For more information see the Darklock and Facehuggershield libraries.


Disc access

Write access for the transformers library has been disabled, preventing it from creating a huggingface cache directory at runtime.

The application itself may still access the disc for reading and writing, however we have restricted reads and writes to the user provided airunner directory (by default this is located at ~/.local/share/airunner).

All other attempts to access the disc are blocked and logged for your review.

For more information see src/security/restrict_os_access.py.


Huggingface Hub

Huggingface Hub contains telemetry and tracking features that have been completely disabled in AI Runner.

The security measures taken for this library are as follows

  • Prevented from accessing the internet
  • Prevented from accessing the disc
  • All environment variables set for maximum security
  • All telemetry disabled

See Facehuggershield for more information.


sudo groupadd docker

For Tasks:

Click tags to check more tools for each tasks

For Jobs:

Alternative AI tools for airunner

Similar Open Source Tools

For similar tasks

For similar jobs