Artists, designers and hobbyists who want a local image-generation app. You need compatible hardware, and some familiarity with image models helps.

What can you build with InvokeAI?

Generate images from text prompts on your own hardware with open models. Fix or extend parts of an image using inpainting and outpainting on the canvas. Build a reusable node workflow for a specific image production task and share it. Compare how different models such as SDXL and Flux respond to the same prompt. Upscale and organise a collection of generated images into boards.

AI-generated illustration

invoke-ai /

InvokeAI

A local web app for making and editing images with Stable Diffusion and Flux models

28.8k stars
TypeScript
Updated today
Beginner friendly
Image, video & audio

Open on GitHub (opens in a new tab) Website (opens in a new tab)

Invoke is an image-generation app you install on your own computer. It starts a local web server and a React interface in your browser, where you write prompts, pick a model and make images. It is free to use under the Apache-2.0 license, which allows commercial use.

Beyond plain text-to-image, it has a Unified Canvas for painting, inpainting (filling in part of an image) and outpainting (extending an image past its edges). It also has a node-based workflow editor, where you connect steps into a custom generation pipeline and share it with others.

The README lists many models: Stable Diffusion 1.5, 2.0, SDXL and 3.5, the Flux family, Qwen Image, Z-Image and more. A few, such as Nano Banana, GPT Image and Wan, work through an API only. It also includes a model manager, a gallery that keeps the prompts and settings behind each image, and upscaling tools.

What it does

Unified Canvas
A canvas with brush tools, inpainting and outpainting. You can use it to extend or fix AI images, sketches, photos or renders.
Node-based workflows
Build custom generation pipelines from connected nodes through a UI, then save and share them for repeatable work.
Wide model support
Works with SD 1.5, SDXL, SD 3.5, the Flux family, Qwen Image and others. It reads ckpt, diffusers and some gguf model files.
Gallery and boards
Images are organised into boards and carry metadata, so you can recall the prompt and settings behind any result.
Model and embedding managers
Built-in tools for managing models and embeddings, plus upscaling and SAM/SAM2 object segmentation and selection.

What you could build

Generate images from text prompts on your own hardware with open models.
Fix or extend parts of an image using inpainting and outpainting on the canvas.
Build a reusable node workflow for a specific image production task and share it.
Compare how different models such as SDXL and Flux respond to the same prompt.
Upscale and organise a collection of generated images into boards.

Get started

Download the Launcher from the project's releases page and use it to install Invoke on your computer. The README points to the installation docs and FAQ at invoke.ai for help.

Commands are copied from the project's README. Read it on GitHub for the full setup.