Quick Start Tutorial

Are you having trouble with activation?  Read this troubleshooting guide

Beginner’s Lesson: Create an SDXL Fat Cat

1 minute

Prep: First, login to your Graydient Account. Everything is preinstalled on our AI cloud. You can do this on any PC or smartphone.

Already an Expert?

If you’re already familiar with Stable Diffusion, you can disable Easy Mode

Easy mode has the “brew” button. It looks like this:

How to use Brew

Take a moment to review the elements of your work space.

  • There’s an input field for your text, called the “positive prompts” box.  You can describe any image that you want to create here. It helps to move the most important words to the front
  • A “recipes” carousel with art styles.  If you don’t choose a recipe, a random one is elected.
  • The brew button (bottom right), which activates the recipe

If you don’t choose a recipe, one will be selected at random

Some recipes are very opinionated, so if your image is very different from what you asked, choose a different recipe.  There’s hundreds in the system. For more control, use the Advanced Mode

Let’s make that fat cat using the SDXL recipe

 

Step 1:  Copy this prompt

Figure (A) above: The first input box is called a “positive prompt”. When writing your positive words, put the most important subject matter first, like this:

a beautiful gray chubby cat in a garden

Step 2: Choose SDXL from the recipes

They’re organized alphabetically.  Scroll the carousel until SDXL appears, and click on that.

Step 3: Click Brew

And you’re done, it’s that easy.

Tip: You can also call recipes using a hashtag, like #sdxl.  You don’t have to scroll and pick it, if you know the recipe name

 

Easy mode is a ton of fun, but there’s a lot more power under the hood.  If you have 1 more minute, try this next prompt.

 

Intermediate Lesson: Create a fancy fat cat

The more descriptive you are, the better your results will be.  Copy this tricky prompt, created by one of our community members, Tomato:

((best quality,highres, masterpiece, vibrant)) by Julio Larraz, big garden, beautiful gray chubby cat, British short hair, plants, peonies, liminal, patterns, big room, free space, intricate nature, additional details, realistic cat (8K UHD, hyper realistic,highly detailed,intricate detail,perfect lighting,award-winning photo,dynamic,detailed background,vibrant colours,background details,reflections,ultrasharp focus,crisp edges,perfect image) [[low resolution, worst quality, blurry, plain background, white background, simple background, normal quality,bad quality]] #sdxlreal

What’s different:

  • #sdxlreal is a recipe, a quick shorthand. It can be selected from the menu, or typed to save time
  • Positive words are nested in round brackets (like this).  Each pair of nested round brackets adds 1.1x emphasis
  • Negative words are nested in square brackets [like this].  Nesting more subtracts 1.1x emphasis per

 

Positives and Negatives

Positive prompts and Negative prompts are words that tell the AI about what we do and don’t want to see. Humans don’t typically communicate this way, so it may seem a little unusual at first, but in a very noisy environment we might say “not that, but this!” and that’s essentially what we are doing. Literally ((give me what I want)) [[but, not, these, things]].  ((pretty cat)) [[ugly cat]].  You want a (cat) but not a [dog].  This prompting style gives you both power and precision.

Another example: Imagine trying to order a chocolate donut in a very noisy, shouty cafeteria. Talking to an AI is a little like this.  You can ask for a chocolate donut, so it’s so much clearer if you say that you want a ((chocolate donut)) [[strawberry, vanilla, almond]]. It really drives the point home.

While it may seem redundant, you can also say ((give me good pictures)) [[but not bad pictures]] every time.  There are better ways to express this than typing out such things every time, which we’ll get to in the Embeddings tutorial.

Summary

You understand that choosing a specific art style is possible, and that there are positive and negative ways to reinforce the AI to understand you better. Have fun making a few more images, and try the next lesson when you’re ready.

Return to the main menu

 

Glossary of Terms

AI model – Also called a concept, an AI model is a new paradigm in computing. Think about it as not quite a database nor a collection of pictures, but a file that contains learned patterns. For example, a model could be trained to learn what oil paintings, horses, or astronauts look like in general and then produce them in various situations. Stable Diffusion is a popular model.

Prompt – The instructions sent to the AI are called prompts. The above is an example of a Stable Diffusion prompt, with some formatting short-hand codes unique to our apps. Each word that an AI recognizes is called a Token

( ) – Round brackets make words Positive. It means that something is important, they create emphasis. Every nested pair of ( ) increase the factor by 1.1x.

[ ] – Square brackets de-emphasize an idea.  Adding [[low quality, worst quality]] results in a higher quality picture. Each pair reduce the factor by 1.1x.

Be diligent about counting and closing your positive and negatives, and never mix the two inside each other.

# – Hashtags invoke recipes, which are community-created art style macros. In this lesson, we tried #sdxl and #sdxlreal.  Also try #boost and #everythingbad, some of our favorites. You will learn how to create your own recipes in a future lesson.

 

Prompting tips:

  • Word order matters. A good general structure for prompts is “((1-3 tokens that boost quality)) A subject that looks like this, doing this action, some other characteristics, at this place [[then your negatives go last]]”. However, some people prefer to load negatives first. There’s no hard rule, but we find that this framework can be most helpful for beginners, as it’s easy to remember and is effective.
  • Put the most important words towards the start. Overdoing the emphasis like (((nice smile))) and [[[teeth]]] can glitch the photo if it can be interpreted as conflicting information. It’s never good to get into a shouting match with your prompt, instead try reordering what’s most important first. If blue squares appear, think about this. You’ll learn about a parameter called Guidance (aka CFG) in a future lesson that can also help you mitigate this.
  • One recipe works best – It is possible to load multiple recipes, but glitches and conflicts may occur.

FAQs:

The #1 question we get: “Can I make nudes, or will I get banned?”

You will not get banned, unlike other services, nudes on here are fine. We have published clear content guidelines, with specific examples.We only ban someone if they’re doing something that’s against the law or bothering other customers.

We also don’t look at your private images, but there is automated software that will alert a moderator if the prompts are suspicious. We try not to ban people lightly, you really have to really screw up to get kicked out of our community.

 

Are there any technical requirements or hidden fees?  

There are many ways use StableDiffusion, but we found all of them overly complicated, required high-spec expensive hardware costs up-front, and unfortunately many low-cost options are full of hidden microtransactions to use the full power of the service, so we wanted to make something better that we would also enjoy using.  We’re huge AI fans ourselves.

With Stable2go, everything is preinstalled and it works fast on every device — a gaming computer or the latest GPU or iPhone are not required at all.  You can create images within seconds on your ten year old computer with our software.

 

How is this different from other AI generators?

It’s our own software. Our team is also unique: we’re the same people that created the popular gaming community Destructoid.

This isn’t a fly-by-night AI generator. We are building a large set of tools and a real community of AI fanatics, like us. We have also regularly sponsored AI model creators behind community favorites like Hassaku/GrapeFruit, Duskfall, URPM, DreamShaper, Majic, and are patrons and investors of other cool AI projects. We also built an API, so other developers can use our general technologies. We are by enthusiasts, for enthusiasts.  Our Telegram sister app is called PirateDiffusion.

Unlike other AI products, the main backbone of all of our apps is an our exhaustive library of selectable models and community-made templates. At the time of writing this tutorial, you can choose over 3,000 fine-tuned AI models to give you the exact look, detail, pose, and lighting that you want, every time.

 

May I make a suggestion?

Absolutely!  We would love your feedback – join our public roadmap and upvote suggestions, and add your own.

 

Back to Web tutorials   |   Telegram tutorials


Terms of Use   |   Privacy Policy   |   Powered by Graydient Platform