zylarian/dockyard-invokeai repository overview

⁠InvokeAI for Dockyard

Leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media.

⁠Quick Start

make service invokeai start

Access the interface at http://localhost:9095⁠

⁠Configuration

⁠Environment Variables

⁠GPU Support

This service involves heavy GPU usage. It is configured to use NVIDIA GPUs by default. Supported devices: cuda, rocm, cpu (slow).

⁠Setup

When you first start the container, it may take some time to download default models. Check logs with make service invokeai logs to see progress.