InvokeAI

Open-source studio for Stable Diffusion with a canvas-based workflow

AI Image GenerationFree planOverseasβ˜…β˜…β˜…β˜…β˜† 4.0

What is InvokeAI?

InvokeAI is an open-source creative studio for running Stable Diffusion and other open image models, aimed at people who want more structure than a prompt box but less tinkering than a node graph. It centres on an infinite canvas where you generate, mask, inpaint, outpaint and composite, with layers and control layers for depth, edges and poses. It runs locally on your own machine or a server you control, keeps a searchable gallery of every generation, and supports multiple model families. Individual use is free; a hosted version exists for teams that need shared infrastructure and review.

Last updated: 2026-09-20. This site only provides an index; for exact features, pricing, and licensing, see the official website.

Key features

  • Infinite canvas for generating, masking, inpainting and compositing
  • Control layers for depth, edges, pose and other guidance inputs
  • Local or self-hosted execution using your own graphics hardware
  • Searchable gallery that keeps metadata for every generation
  • Support for multiple open model families, LoRAs and embeddings
  • Unified canvas and node editor inside one application

Pros & cons

Strengths

  • Open source and fully self-hosted, so nothing leaves your machine
  • The canvas workflow suits compositing better than a prompt box
  • Every generation keeps its settings, which aids reproducibility

Watch out for

  • Installation and model management need some technical comfort
  • Requires a capable GPU for a smooth working experience
  • The interface has a learning curve compared with hosted apps

Best for & use cases

local image generation, inpainting and compositing, iterative art workflows and privacy-focused use

If you're comparing similar products, check the alternatives below, or browse all tools in the AI Image Generation category.

FAQ

Is InvokeAI free?

The application is open source and free to run on your own hardware. There is also a managed cloud offering aimed at teams, which is paid and adds sharing, review and hosted compute.

What hardware do I need?

A modern GPU with a decent amount of video memory makes the biggest difference, and the project publishes guidance for different setups. CPU generation is possible but slow.

How is it different from ComfyUI?

Both run open models locally. ComfyUI is node-based and extremely flexible, while InvokeAI favours a canvas and layer interface that feels closer to a traditional image editor.