InvokeAI Download: A Professional Canvas for AI Images
InvokeAI 6.14.3 · for Windows, macOS and Linux
Developer: Invoke (opens in a new tab)
- Windows
- macOS
- Linux
The download link points to the program’s official source; the file is not hosted on this site. Link address: github.com
Link not working? Let us know
InvokeAI (Invoke) is a free, open-source creative engine for diffusion models with a web-based interface, a unified canvas and node workflows. The latest stable version is 6.14.3, released on October 29, 2026; it installs with the Invoke Launcher on Windows, macOS and Linux and is licensed under Apache 2.0.
InvokeAI is built for people who treat AI image generation as part of a creative workflow. Its Unified Canvas lets you paint, extend and refine images in layers, and it also serves as the foundation for several commercial products.
What does InvokeAI do?
- Unified Canvas: Inpainting and outpainting with brush tools on a single canvas.
- Workflows: A node-based editor for building and saving your own generation pipelines.
- Gallery: Boards and a gallery that remember the settings used for each image.
- Wide model support: SD 1.5, SD 2.0, SDXL, SD 3.5 and several FLUX variants, among others.
- Extras: Upscaling, model and embedding managers, and SAM/SAM2 segmentation.
Who is InvokeAI for?
It suits artists and designers who want polished editing tools, and anyone who prefers a tidy interface over a maze of settings.
Installing with the Invoke Launcher
Get the Invoke Launcher for your system from its GitHub releases, run it and let it install Invoke. The launcher installs Python and the dependencies for you. Advanced users can install manually; that route requires Python 3.12, as newer versions are not supported yet.
Hardware requirements
Invoke runs on Windows 10+, macOS 14+ and Linux (Ubuntu 20.04+ is well tested). Minimum NVIDIA requirements for the most common model families:
| Model | GPU | VRAM | RAM |
|---|---|---|---|
| SD 1.5 | GTX 16xx+ | 4 GB | 8 GB |
| SDXL | RTX 20xx+ | 8 GB | 16 GB |
| FLUX.1 | RTX 20xx+ | 10 GB | 32 GB |
- NVIDIA: Compute capability 7.5 or newer (GTX 16xx, RTX 20xx and later). GTX 9xx/10xx cards are not supported by the standard CUDA build.
- AMD: Windows and Linux x86_64 through ROCm, with the same VRAM needs as NVIDIA.
- Intel Arc: Windows and Linux x86_64.
- Apple Silicon: All M-series Macs work; 16 GB or more of memory is recommended.
Cards with less VRAM often still work, only more slowly.
License
InvokeAI is free and open source under the Apache 2.0 license, which allows commercial use. Models have their own licenses.
Pros
- Polished canvas for painting and editing
- Node workflows for repeatable results
- Easy install with the Invoke Launcher
Cons
- Newer models need a lot of VRAM and RAM
- Older NVIDIA cards (GTX 9xx/10xx) are not supported
Frequently Asked Questions
How much VRAM does InvokeAI need?
For SD 1.5, 4 GB of VRAM and 8 GB of RAM; for SDXL, 8 GB of VRAM and 16 GB of RAM; for FLUX.1, 10 GB of VRAM and 32 GB of RAM, according to the official requirements.
Can I use InvokeAI commercially?
Yes. InvokeAI itself is under the commercially friendly Apache 2.0 license. Check the license of each model you use separately.
InvokeAIOpen source
Download (opens in a new tab)
