How to publish your style model to the exactly.ai style library

Publishing your style to the exactly.ai style library is a great way to increase the visibility of your artwork and grow your platform as an artist. In contrast to our private models which are accessible only to you, public models are published on the platform and accessible to everyone.

Why publish your style?

By publishing your style, your model will feature in our community library alongside other professional creators. Your style will be accessible to the public and everyone, including our community of one hundred thousand artists, designers and brands can access your model and generate images with it.

Plus, once your public style gets verified you will be able to sell commercial licences for images generated with it. This means you’ll earn royalties each time someone downloads an image generated with your model for use in commercial projects. To learn more about image licensing on exactly.ai read our blog post.

Although your style will be available to the public for use exactly.ai ensures that all the artwork generated with it remains under your copyright. No matter how many times people generate and experiment with your model, the copyright of the model, the training images and the images created belongs entirely to you. You can read more about our legal principles here.

In this post, we’ll focus on how to compile the training images (dataset) you’ll use to create your style model and how to submit your private style model to our style library. Read this blog post for more information on how to create a stylish AI model.

When compiling your dataset there are three key things to keep in mind:

1. Image size matters

Your AI model will learn to replicate whatever it’s been trained on. To ensure that it generates only the highest quality artwork, you’ll need to use professional, high-resolution images in your dataset.

Here are the details:

• Minimum size: Your images need to be at least 1024 x 1024 pixels if they're square. If you’re using a images of a different shape, the total area needs to be the same as a 1024 x 1024 image. For example, a 16:9 image should be at least 1376 x 768 pixels.
• Don’t upscale: Don't try to cheat the system! Upscaling a smaller image to 1024 x 1024 won’t work. Your images need to be genuinely high-resolution even if it means re-exporting.
• Aspect ratio: Stick to a consistent aspect ratio (the relationship between width and height) of 16:9 or 9:16. While you technically can go up to 1:2 or 2:1, it’s best to avoid that for now.

2. Consistency is key

One of the most important things to remember when compiling your images is consistency. Your dataset should have a strong, unified style.

Your images should reflect your unique artistic style, not just a general technique. For example, don't upload a collection of images that are all line art or pop art. Instead, make sure they have your specific take on line art, your individual approach to pop art, or your unique style of digital painting.

Here’s an example of a perfectly consistent dataset. This artist has a clear signature style that is visible in every image.

Mixing styles is a big no-no!

If your work encompasses a number of different aesthetics, train a different model for each one. Even if it means excluding parts of your portfolio when training, the final model will have a stronger, more consistent style which is crucial for making your work commercially accessible on the marketplace.

You can't train a model to take the style from one set of images and the anatomy from another. It just won't work.

Here’s a dataset which lacks consistency. This artist is experimenting with different styles, which will confuse the AI model.

3. Variety is the spice of AI

While consistency in style is crucial, you need to demonstrate that style across a variety of different subjects.

Think of it like this: the aim of your model is to learn how your style applies to new objects which you haven’t trained on. This means it needs a rich understanding of how your unique style applies to the world so it can generate more creative and diverse outputs to new objects.

Instead of showing your model 20 pictures of a cat in different poses it’s much better to show it 5 different animals, 5 different people, 5 different objects, and 5 different landscapes. Otherwise your model becomes fixated on cats and limits its commercial viability!

The Golden Rule: Aim for at least 2-3 images from each of these categories (the more, the merrier!):

• People
• Animals
• Nature
• Weather
• Food & Drink
• Travel & Places
• Objects

You can focus a bit more on one or two categories, like people, nature, or animals, but don't overdo it. Avoid uploading only people, only flowers, or only objects. Keep it balanced!

Why is this important? A diverse dataset helps your AI model understand different subjects, styles, and compositions. This leads to a more versatile and powerful model that can generate a wider range of creative outputs. Ultimately, this makes your model able to be used in a variety of circumstances which will attract more clients.

Think of your dataset as a visual representation of your worldview. By including diverse subjects, you're showing your model how you see the world — what you find beautiful, interesting, and worth capturing. This not only makes your model more well-rounded but also allows it to generate images that are infused with your unique perspective.

Here’s a great example of a diverse dataset: it includes people, nature, objects, and animals.

And here’s a dataset which lacks diversity. It only features portraits.

To sum up

To create a public style which passes our review process and will be shared with the exactly.ai community, your dataset needs to meet these essential criteria:

• High resolution training images: all training images must be at least 1024x1024 pixels, and remember, the higher resolution the better!
• Consistent style: your unique artistic style should shine through in every single image.
• Diverse subjects: include at least 15 training images (and ideally more!) from a variety of categories, like people, animals, nature, and more.

By following these guidelines, you'll be well on your way to creating a successful and engaging public model that showcases your artistic talent and inspires others!

And finally, here’s an ideal dataset of 24 images

Training and submitting your model

Once you've assembled a dataset that meets the criteria outlined above, it's time to train your style model and submit it for review! We cover the ins and outs of training a model in this guide.

When your model is ready, you can send it for moderation — in the sub-menu form the model library choose the option “Publish model”.

Then, sit back and relax while we review your submission. We'll email you to let you know whether your model has been accepted or rejected.

If your model is accepted, congratulations! We'll take care of the rest and prepare your model to be published on the platform.

If your model is rejected, don't worry, you can always retrain your model for free and resubmit it!

Your model might not be approved for a few reasons:

• Terms of use violation: your training images do not align with our terms of use.
• Inconsistent style: your model may not have been trained with a consistent artistic style.
• Copyright issues: we may have concerns about the copyright of the images used.

Before you submit, don't forget to set up your profile!

This is your chance to present yourself to the exactly.ai community. Once your model is published, your profile information will be displayed alongside it allowing people to find your work and grow your following.

What happens next?

Congratulations! You've successfully published your AI model on exactly.ai. Now what?

The next step is to turn your model into a verified public model and start earning royalties from it!

Firstly, make sure to share the good news on social media, this helps attract attention to your model and get your style verified.

Once 25 users have tried your model, you'll receive an email with the opportunity to verify the copyright status of your images. A verified public model allows subscribers to use generated images for commercial projects, and you, as the creator behind the model, will be paid for it.

For now, that’s everything you need to know to create a public model. We’re excited to have you on board along with your work. Start training your style today and get your work seen far and wide — just as it deserves to be!

Sound exciting? Sign up now to publish your first public model.