itch.io

The extension probably won't be able to do sprite sheets any time soon, but on the website (https://www.retrodiffusion.ai/) we've nearly got a model released that can do walk cycles like shown here: https://x.com/RealAstropulse/status/1896924271659884854

Yep! You can access those models from inside the extension through "help" -> "retro diffusion tools" -> "download models"

It is not supported by AMD's AI drivers in pytorch

I haven't heard of this issue before, would you join the community discord server so we can figure out whats going on more closely? https://discord.gg/retrodiffusion

It's a way of controlling how the image generates using some input images. You can read about it here: https://stable-diffusion-art.com/controlnet/ (This is about general controlnets and how they work)
Or you can see a video I did going over how they work in Retro Diffusion here:

Hey! There can be false detections sometimes because of how windows sorts graphics outputs, especially if your CPU is amd and has integrated graphics.

The comments here isn't a good place for troubleshooting, so if you'd join the community discord server and make a post in the errors channel that would be great: https://discord.gg/retrodiffusion

It creates something new, this model doesnt have the consistency to make the same characters in different poses.

You can make animations by hand in Aseprite, but Retro Diffusion can't generate animations or animated spritesheets.

You would need to use Linux with an AMD gpu. Additionally only more recent AMD cards are supported, 7000 series and up, with the exception of the 6700 and 6800 which are also supported.

It cannot make animations yet, though it is something we are working on.

No, it runs locally on your own hardware. It will only stop working if you delete the files or other things it relies on.

Looks like Python wasn't able to install the libraries it needed. Being an AI program it's incredibly complex, but I do my best to simplify it down for users. Unfortunately sometimes it needs a little help to get installed and running.

Itch.io comments aren't the best place for troubleshooting, so please join the community discord server so I can help you with this better: https://discord.gg/retrodiffusion

Yep, it can do any aspect ratio you want. Some might look a bit weird, but anything less extreme than 1:3 should look fine :)

Hey! It's not specifically trained for this style, so it won't do it very well.

The program doesn't have any animation tools, but I'm wondering what you mean by a japanese style? If you mean something more like a 'chibi' style you can get this through prompts, without the need to use the modifier models.

I'd love to help it meet your needs- RD is quite flexible, but it does require some tinkering and learning to use effectively. Of course if you'd rather just refund, you can contact itch.io support for that.

No, the cloud version still requires you to purchase credits. The hardware requirements are listed very clearly on the product page.

Hey! You do need Python 3.11 (3.11.6 is the best specific version). It is unable to use python 3.12 at all, because the largest dependency is PyTorch, which is only available for <= Python 3.11

Yep, it runs completely locally. All the artwork used for training was either made by me, or was given with consent from other artists.

Low chance, but Pixelorama has been on my radar for a while :)

You can run text to image and image to image, but Neural tools and controlnet aren't compatible with cpu mode yet. I'm working on it, but no release for it right now.

The issue is that your GPU is not compatible with the software. AMD can only be used on Linux. This is explained in the compatibility section.

Yes! Absolutely. I included that info in the download section of the gumroad page, but forgot to put it in this page :)

Hey! Thanks for the review!

Here's some tips for tiling and portraits: 

With tiling, there are two options that do different things. The "Tile X direction" and "Tile Y direction" can be made visible by toggling "Show advanced options", these settings mathematically force the image to tile. This can be great when you need something to tile that normally wouldn't, but its also very heavy handed and can make weird stuff.
The second option is to enable the "Tiling" modifier. This is a specialized image generator that has been trained on tiling textures, stuff like Minecraft blocks. It works best at 16x16 or 32x32.

For portraits, typically putting something like "A close up portrait of ___" or "A headshot of ___" will do the trick. For example here is "A close up portrait of a fox monk" with all other settings at default values:

On speed and CPU usage:

AI image creation is actually one of the most complex tasks computers can do, when you boil it down, its essentially solving differential equations with billions of inputs and outputs. Way more demanding than any AAA game, or even most 3D rendering software, even for small pixel art images. This is why the compatibility section is so strict.

We've managed to get the requirements down to a 4GB GPU, which is pretty impressive given the size of the models and the complexity of the computations. You can find system requirements and some benchmark data above on the main page.

Again thanks for the review, and I hope it helps people with the decision to buy or not!

Send me a message on Discord with proof of purchase and I'll send you a code to get the model files :)

There is a sale going on right now!

Yep! There is an isometric model in the "Modifier" section :)

https://imgur.com/a/t17Azsa

The model files required by the extension are downloaded automatically, but these files are only usable inside of Aseprite. If you need model files for use outside of Aseprite you can either purchase the extension through Gumroad, or contact me with proof of purchase and I'll send a redeemable code.

I've now included the model files in the program itself, to get them you just need to go to "Help" -> "Retro Diffusion Tools" -> "Download Models"

You can literally read my post above, stating that I had permission for each image we used in training.

4gb is on the low end, but you will be able to make 64x64 images.

All of the training data we use is given to us by artists for the purpose of training, a plurality of the dataset is even my own artwork.

Every time an image is generated it cost us on the server-side, even if no settings were changed, so we can't make that an option unfortunately. Sorry you didn't get what you were expecting. What were you trying to create?

Thank you! You should check out the site, its a bit more affordable and you can even try it for free :) https://www.retrodiffusion.ai

Hey! The paid version does run locally. The goal is for the web and local versions to maintain as many identical features as possible, right now the web is slightly behind but we're updating it this week.

There aren't any tokens with the local version.

16GB RAM is a minimum for CPU only generations. If you have a capable GPU you only need 8GB of RAM.

There isn't an explicit refund policy, but we provide refunds to anyone who asks.

The pixel art image size for a 512x512 generation is 64x64. Both the extension and the web version handle all the size conversions behind the scenes for you.

All versions have an image input option.

Based on the hardware you mentioned, you would not be able to run it. Intel iris Xe is not a dedicated gpu, but rather integrated graphics on your cpu. You must have a dedicated GPU with at least 4GB of VRAM. You can refer to this system requirements chart for more detailed compatibility requirements:

There haven't been any updates to the core models since then. When I do change those, they will get updated :)

I've actually been training and releasing models for about 10 months now, all trained on my own artwork and other licensed assets, nothing to do with Pixel Art Diffusion (though KaliYuga is a friend and inspiration). You might actually be interested in this downscaling algorithm I designed myself, its not quite as good at turning things into pixel art as Pixelization, but it is MIT licensed. https://github.com/Astropulse/K-Centroid-Aseprite

I only noticed because I considered using the model myself, until I saw the licensing. Best of luck!

Hey! I'm concerned that the new AI deep pixelization tool is based on this code and model: https://github.com/WuZongWei6/Pixelization

If this is the case it would be a violation of the non-commercial license unless you have arranged a deal with the author.

Is this the case?