Run Powerful AI Models Locally

Chat with LLMs, generate images, convert text to speech, and more — all on your own hardware. No cloud. No subscriptions. Complete privacy.

🔒 100% Private* ⚡ GPU Accelerated 🐧 Linux & Windows 🆓 Open Source
📦 Install Now View on GitHub →

Everything You Need for Local AI

A complete suite of AI tools running entirely on your hardware, with no data ever leaving your machine.

💬

Chat with Local LLMs

Run Qwen, Mistral, Llama, and other models locally. Enjoy fast, private conversations with AI assistants.

🎨

Image Generation

Create stunning images with Stable Diffusion and FLUX. Text-to-image, inpainting, and outpainting supported.

🔊

Text-to-Speech

Convert text to natural-sounding speech with multiple voice options. Real-time voice synthesis.

🎤

Speech-to-Text

Transcribe audio to text with high accuracy. Perfect for dictation and accessibility.

📚

RAG & Document Q&A

Ask questions about your documents. AI reads and understands your PDFs, ebooks, and web pages.

🤖

Agentic Tools

Web search, file operations, deep research mode, and more. Let AI help automate your workflows.

See AI Runner in Action

A clean, intuitive interface designed for productivity and creativity.

AI Runner Chat Interface

Chat Interface

Have conversations with local LLMs in a clean, familiar interface.

AI Runner Chat Interface

Image Generation

Generate stunning images with Stable Diffusion (SDXL), FLUX 1.0 and Z-Image Turbo models.

AI Runner Chat Interface

Nodegraph Workflows

Create complex AI workflows visually with our nodegraph editor.

AI Runner Chat Interface

Settings & Models

Configure models, adjust parameters, and customize your experience.

Get Started in Minutes

Choose your preferred installation method and start using AI Runner today.

🐳 Docker Installation (Recommended)

Prerequisites

Ensure you have Docker and NVIDIA Container Toolkit installed:

# Install Docker (if not already installed)
curl -fsSL https://get.docker.com | sh

# Install NVIDIA Container Toolkit
distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
sudo apt update && sudo apt install -y nvidia-container-toolkit
sudo systemctl restart docker

Step 1: Clone Repository

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

Step 2: Run AI Runner (GUI Mode)

# Allow X11 access and run with GUI
xhost +local:docker && docker compose run --rm airunner

Alternative: Headless API Server

Run AI Runner as a headless API server on port 8080:

docker compose run --rm --service-ports airunner --headless
🔌 API Endpoints: The headless server exposes:
GET /health - Health check
POST /llm - LLM inference
POST /art - Image generation

🐍 pip Installation (Linux/Windows)

Prerequisites

Ensure you have Python 3.10+ and pip installed. CUDA toolkit recommended for GPU acceleration.

Step 1: Create Virtual Environment (Recommended)

python -m venv airunner-env

# Linux/macOS
source airunner-env/bin/activate

# Windows
airunner-env\Scripts\activate

Step 2: Install AI Runner

pip install airunner

Step 3: Download Models

airunner-setup

Step 4: Run AI Runner

airunner
💡 Tip: For GPU acceleration, ensure you have CUDA drivers installed and use the appropriate PyTorch version for your GPU.

💻 From Source (Development)

Step 1: Clone Repository

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

Step 2: Create Virtual Environment

python -m venv venv
source venv/bin/activate # Linux/macOS

Step 3: Install Dependencies

pip install -e ".[all]"

Step 4: Download Models & Run

airunner-setup
airunner
🔧 Development: See our Wiki for development setup, contributing guidelines, and architecture documentation.

System Requirements

AI Runner requires an NVIDIA GPU

💻 Minimum

  • OSLinux (Ubuntu 22.04+)
  • CPU8+ cores
  • RAM32 GB+
  • GPUNVIDIA RTX 5080+ (16GB+)
  • Storage100 GB SSD

🚀 Recommended

  • OSLinux (Ubuntu 22.04+)
  • CPU8+ cores
  • RAM64 GB+
  • GPUNVIDIA RTX 5090+ (16GB+)
  • Storage200 GB SSD