Krita AI Diffusion - Generative Image AI For Krita is Seriously Useful, Powerful and Free (If You Can Install it Locally)
Generative AI sequence of a woman in a business suit. From sketch to refined image using Krita AI Diffusion - by TET
Generative image AI, where you describe an image with a text prompt to an Artificial Intelligence model and it produces a new image based on your prompt, is gaining a strong hold as a tool for many artists.
Krita AI Diffusion brings generative AI image tools right into your favourite free and opensource, graphics editor, Krita. Not only that, if you have a computer with decent specs (and at least 10GB of hard drive space), Krita AI Diffusion is completely free.
What If I Don't Have a Powerful Computer?
If you're in my situation, with a computer that was around before anyone in the mainstream had even heard of generative AI, you can still access Krita AI Diffusion for free, using a cloud based AI server, Interstice and 300 tokens, to get you started.
Once your initial tokens run out, purchase 5000 more for 10€ (approx US$11.00). Tokens never expire. I would've like to have seen a daily quota of free tokens for casual users, like most AI image generation sites, but at least it's not a subscription. You only need to purchase tokens when you run out.
This was my very first image generation, one of my benchmarks of an alien wearing a business suit, sitting at a desk, using a computer. The floating docker on the left is the Krita AI Diffusion Plug-in.
Installation
I'm not going to go through the process of installation as all that is clearly explained on the plugin website. There's even a visual guide on the Interstice site.
You have three options. Two of them involve running a local server on your machine for the AI. You may eventually want to do that but I recommend you take advantage of the 300 free tokens and the cloud server option to try it out.
Setup for the cloud, is as easy as downloading and installing the Krita AI Diffusion plugin (a very small zip file) and then linking your free account on the cloud server to the plugin. You can be up and running inside 15 minutes.
Note that no images are stored on the server and you own all the rights to the images you generate.
Using Krita AI Diffusion
Krita AI Diffusion installs as a docker panel inside Krita. Once you've started a new document and connected it to your server (in my case the cloud). You're good to go. Type in a prompt and see what happens.
The main AI features you get are Generate, Inpaint, Outpaint, Refine, Live Painting, Control (guided output from image references), Resolutions, Job Queues, History, Style Presets, and Style Preset Customization.
I was both impressed and overwhelmed with what this plug-in brings to Krita. On its own it seems like one of the most powerful AI image generators out there, but teamed with Krita, a full feature graphics editor, I almost don't want to use anything else. It's kind of next level.
If you are using the free tokens of the cloud service I recommend you keep your image size to 512 x 512 pixels. You can go bigger but you'll burn through your tokens faster (the number of tokens per generation is shown next to the 'Generate' button based on your settings).
Krita AI Diffusion's live canvas lets you paint fairly basic drawings which it will turn into more finished art taking your text prompt into account.
Also, avoid using the live canvas feature too much while using your free tokens, as that seems to burn through tokens extremely quickly. My Interstice account said I'd generated 121 images, which I definitely didn't. I can only assume, every time the live canvas updated, that also counted as a generation.
There's too many features to actually cover in an article but if you want to see Krita AI Diffusion in action watch the introductory video - part 2 below by Intelligent Image for a fairly comprehensive overview. Note that part 1 to this video is a walk through of installation and all the plug-in's settings that is worth watching if you want to know what everything goes.
One feature I do like is the ability to have different image generations on each layer of a document. This allows for so much more flexibility as you're not limited to just changing a single layer like most online generative editors.
If I didn't have to pay for tokens I would spend a few weeks learning everything. It's a real game changer for Krita itself as a graphics application. That said, 5000 tokens goes a long way given a standard image generation is between 1-5 tokens.
Final Thoughts
If you use Krita, I highly recommend installing this plug-in and at least trying out the cloud version for free until your tokens run out. If it's not for you, just turn off the docker and never look at it again.
However, what it brings to your workflow is truly astonishing. Aside from the obvious, of completely generating images from scratch, it can be a lot of fun taking your own sketches, that you were never going to do anything with, and spending a few minutes using AI to color and enhance them to finished, digital art.
It's also very good at turning images back into line art. Very handy if you've ever thought of making coloring books of your own artworks.
I can't even begin to sing its praises enough here. I would encourage you to watch as many videos as you can on Krita AI Diffusion to see what is possible. It really is one of the most powerful AI Image Generators out there.
o---o--- ---o--- o---
Did you find this article useful? Subscribe to my newsletter and get the latest articles delivered to your inbox.
M obile productivity apps have become so capable that they can be great alternatives to their PC/MAC equivalents or serve as great tools in their own right when you're away from your desk. While some apps simply mimic their desktop counterparts, others offer well thought out, touch-friendly interfaces that are easier and more fun to use. Every so often I check out what's available for 2D animation for Android devices, since that's what I use, that can complement my workflow with Reallusion's Cartoon Animator 5. Some may be available for Apple devices as well. Below I've listed six free (F) apps (with optional paid (P) upgrades) on the Google Play Store that you might want to explore. Some are just fun apps on their own while others may be useful as part of your workflow on bigger animation projects. Not all are exclusively animation apps and could be used on any production. JotterPad (F/P) The name JotterPad makes this sound like a notepad application but it's ...
Scene 2 from my storyboard, Infinite Monkeys. Work on my Infinite Monkey animated short is definitely back on track since my last update and my previous part 4 post . To get things moving again I've decided not to spend time creating a final draft storyboard that uses all the actual props and characters. This is something that you'd generally do if you were planning to show the storyboard to someone else for feedback. At this point I'm not planning to do that so I'm going bypass it and create and animate each scene as I go, saving a bit of time.
In part one of my deep dive into the free VTuber software, Inochi2D , I focused mainly on Inochi2D Creator, which is used for rigging your character avatar in the correct file format for use with Inochi2D Session, the puppeteering part of the software. The two sides of the software are still very much in development and the documentation, particularly for Session, is very thin on the ground. To the point where I don't think I could even do a comprehensive tutorial because I'm not sure I'm even doing things right, and the software could change significantly in a single update. As a result, in this part of my Inochi2D deep dive I'm changing tact from presenting my finished Cartoon Animator TET Avatar, and will be summarizing my experience of getting Session up and running using OpenSeeFace as the recommended webcam motion capture software. To do this I will be using the TET avatar I created in my review of Mannequin , since that can be exported as a full, ready to go r...
Random character head, Biro sketches drawn by TET (left). Render by PromeAI (right) using Prome's Sketch Render tool set to 'Comon:Cartoon, Render Mode: Outline'. W hile I don't do New Year Resolutions, one of my plans for the year ahead is to do more of my own art. Specifically character design drawn in an actual, physical sketchbook. To that end, I have been spending the last half hour of most days drawing a page or two of random biro sketches in my sketchbook and posting the pages to my Instagram account (this link will take you to one of my posts). These sketches are mostly practicing my skills because I don't really draw regularly anymore. Here is a tip, if you do this kind of sketching, and push yourself to keep doing it, you will see many drawings that could be taken further, even if you don't have anything they're suited for just at the moment. Which is where my second favorite AI Image Tool (after Leonardo.ai ) PromeAI comes into play. PromeAI ...
Wonder Unit Storyboarder. As an independent and solo animator I'm always tempted to try and skip storyboarding my animated shorts because they're usually only single scene sketch comedy type jokes. As a result I have many unfinished projects that kind of petered out due to having no clear finishing line. Storyboarding your productions, no matter how small, gives you a step by step guide of every shot that needs to be completed (no planning shots as you animate). It also allows you to create an animatic that gives you a rough preview of the finished production. In short, you shouldn't skip storyboards as they, generally, increase the chance of the project being completed. Disclaimer - I'm Not a Fan of Storyboarder Upfront, Wonder Unit's Storyboarder is not my preferred storyboarding software. However it's completely free, has a number of very compelling featu...
Young TET VRoid Avatar in VSeeFace. In a previous YouTube video I demonstrated how you could use a webcam, Cartoon Animator and its MotionLive 2D Facial Capture system to create a virtual avatar that you could perform live during a live stream. Unfortunately the motion capture performance of this system is very average to say the least, and is even more expensive if you want to add hand motion capture via a Leap Motion Controller . This lead me down a rabbit hole of seeing what other solutions are out there, where I bumped into the world of the VTuber or Virtual YouTuber (creators who puppet and voice an on camera avatar instead of appearing on camera themselves). Specifically VTuber software, which has been a real eye opener for how expensive and far behind Reallusion's upper body motion capture system actually is. The VTuber world is heavily entrenched in the world of Anime so you'll see a lot of 2D and 3D avatars in this style. If fact, if an Anime avatar is your thing, y...
If you're the kind of 2D animator that needs to get high quality work out fast then you should be taking a serious look at Reallusion's Cartoon Animator 4 because that is the key feature that informs almost everything this application is capable of. Formerly known as CrazyTalk Animator 3, the release of Cartoon Animator 4 (CA4) reveals an application that goes out of its way to make high end 2D puppet rig and motion graphic animations easier to create quickly, and as intuitively as possible. That's not to say there isn't a learning curve but another great strength of CA4 is that you won't get lost in a sea of icons, tools, menus and more. The initial user interface is less complicated and more logically laid out than one may have used to put together a powerpoint presentation.
Very interesting, I am definitely going to have a look..
ReplyDelete