InvokeAI 2.2: UI Outpainting, Embedding Management and more

2 December, 2022

InvokeAI 2.2 has been released:

- An interface for drawing, doodling around pictures, adding new objects inside — I call it the new Photoshop.
- Installers are now automatic, click and it will do everything itself.
- All sorts of different improvements.

Meanwhile:

- Apple has moved SD execution to its AI chip, speeding up the process by half and getting rid of RAM consumption https://github.com/apple/ml-stable-diffusion.
- The SD developer said that the new model is 20 times faster and it takes only 4 steps and less than a second to generate a quality picture.
This means that the picture generators will run locally on almost all devices, including iPads and iPhones, in near real time.

Release Notes:

InvokeAI 2.2 is now available to everyone. This update brings in exciting features, like UI Outpainting, Embedding Management and more. See highlighted updates below, or the full release notes for everything included in the release.

The Unified Canvas: the Web UI now features a fully fitted infinite canvas that is capable of outpainting, inpainting, img2img and txt2img so you can streamline and extend your creative workflow. The canvas was rewritten to improve performance greatly and bring support for a variety of features like Paint Brushing, Unlimited History, Real-Time Progress displays and more.

Embedding Management: easily pull from the top embeddings on Huggingface directly within Invoke, using the embed token to generate the exact style you want. With the ability to use multiple embeds simultaneously, you can easily import and explore different styles within the same session!

Viewer: the Web UI now also features a Viewer that lets you inspect your invocations in greater detail. No more opening the images in your external file explorer, even with large upscaled images!

1 Click Installer Launch: with our official 1-click installation launch, using our tool has never been easier. Our OS specific bundles (Mac M1/M2, Windows, and Linux) will get everything set up for you. Our source installer is available now, and our binary installer will be available in the next day or two. Click and get going - It’s now much simpler to get started.

DPM++ Sampler Support (Experimental): DPM++ support has been added! Please note that these are experimental, and are subject to change in the future as we continue to enhance our backend system.

Up Next

We are continually exploring a large set of ideas to make InvokeAI a better application with every release. Work is getting started to develop a modular backend architecture that will allow us to support queuing, atomic execution, easily add new features and more. We’ll also officially support SD2.0 soon.

If you are a developer who is currently using InvokeAI as your backend, we welcome you to join in on the conversation and provide feedback so we can build the best system possible.

Our Values

With increasing adoption of InvokeAI by professional creatives and commercial projects, we feel it is important to share our values with the community that is choosing to put their belief in our work.

The InvokeAI team is fully committed to building tools that not only push this incredible world of generative art further, but also empower the artists and creatives that are pivotal to this ecosystem. We believe we share a role in developing this software ethically and aim to navigate all community concerns in a meaningful way. To learn more, please see our statement here.