InvokeAI
A self-hosted AI image generator that runs on Windows, macOS, Linux and Docker, with canvas editing, visual workflows and an Apache 2.0 license.
InvokeAI is a free, open-source image generator for artists and production teams who want to create and edit images on their own hardware. Its locally hosted web interface brings image generation, canvas editing and workflow tools into one application. With local models, prompts and artwork stay on the machine you control.
The Unified Canvas lets you paint, sketch, and change parts of an image using layers, brushes, inpainting and outpainting. Layers remain independently editable. ControlNet can guide a generation using poses, edges or depth maps, while batch generation, prompt wildcards and upscaling help with exploring and refining concepts.
For repeatable work, a visual node editor lets you build generation workflows and expose the controls you use most often. A gallery keeps images and their prompt and settings metadata available for later edits. The model manager handles checkpoints, LoRAs, Textual Inversions and ControlNets. Supported model families include Flux, SDXL and SD 1.5; the README also lists support for ckpt, diffusers and some GGUF models.
InvokeAI runs on Windows and Linux with an NVIDIA or AMD GPU, or on an Apple Silicon Mac. Docker is another self-hosted option. The project has an Apache 2.0 license.
