InvokeAI

InvokeAI (now branded Invoke) is an open-source Stable Diffusion suite built for professional creators. Originally created by lstein and open-sourced in August 2022, it positions itself as a category-leading WebUI and the underlying engine for several commercial products. Unlike A1111's hacker-oriented interface, Invoke emphasizes a professional creative workflow. The repository has 27k+ stars on GitHub and was last updated in early July 2026, making it the most actively maintained of the five major local SD WebUIs.

Core capabilities break down into three areas. First, Unified Canvas: an embedded canvas with inpaint/outpaint, brush tools, and layer management that treats AI generation as collaborative painting on a canvas — ideal for illustrators doing iterative local edits. Second, node-based workflows: visual nodes let users assemble generation pipelines, save frequently used flows as reusable templates, and share them with teammates. Third, model management: the built-in Model Manager supports twenty-plus mainstream checkpoints — SD 1.5, SD 2.0, SDXL, SD 3.5 Medium/Large, Flux.1 Dev/Schnell/Kontext/Krea/Redux/Fill, Qwen Image/Edit, CogView 4, Wan, Z-Image — in ckpt, diffusers, and quantized gguf formats. Also included: an Embedding Manager, Upscaling Tools, and SAM/SAM2 segmentation.

Use cases include designers who need canvas-based local iteration, teams that want to capture generation flows as reusable workflows, multi-model users running SDXL, Flux, and SD 3.5 in the same tool, and companies building commercial products on top of the engine (the project notes it "serves as the foundation for multiple commercial products"). The onboarding bar is moderate: an official standalone Launcher downloads the desktop app and handles updates in one click; full documentation lives at the docs subsite; the community is most active on Discord.

Licensed under Apache-2.0, more permissive than A1111's AGPL-3.0 or Fooocus's GPL-3.0, which makes it friendly for teams building commercial products on top.