Master the Perfect Midjourney Prompt Formula (in under 9 minutes)!

there's a better way to structure your AI art prompts to achieve a more controlled and consistent image in this video I'll teach you the four components of prompt structure I'll then do a deep dive into each component with detailed examples and finally I'll go through a comprehensive example to show you how the four components come together and work in tandem you won't want to miss that this prompt structure guide is designed for mid-journey but works with almost all the text to image AI tools let's Jump Right In with the prompt structure overview you can start by breaking your prompt into four separate components marked by the variables x y z and M each letter corresponds to a different component and they are subject style medium and modifiers x marks the spot the main destination the goal and corresponds well to the subject why is the why of the image What feelings does it invoke The Vibes and styles associated with the image next comes Z the well I don't have any witty sayings for this one and GPT wasn't much help I suppose The Zephyr of imagination works either way Z is the medium the art style of the desired image finally we get to M the modifier these are all of the settings related additions you can add just a quick interjection here I've always called these modifiers because they modify the image but mid Journey calls them parameters while editing I realize this might be kind of confusing so just understand that modifier means parent the order in which you include these components in your prompt matters to an extent it's always best to start with your subject and end with your modifiers but the position of your style and medium components are interchangeable I find it best to separate each component with a comma Now it's not necessary to do this but I found it helps me to stay organized it's also crucial to know it's best to stick to one to four keywords for each component mid-journey won't use all of your keywords and it may use your keywords in unintended ways so don't overcrowd your prompts unless you happen to be a chaos enjoyer that's just the overview now let's take a deep dive into each of the different components to really understand how to use them effectively every image has a subject maybe it's a robot or a cat or a zombie or New York City you get the point it's what the image is about and usually it's a noun though it doesn't have to be let's use the example of a cat to illustrate cat is a basic subject and gives basic results even without getting into use of the other components we can refine our subject further by implementing more packed keywords I'll have a tutorial on packed keywords out soon but a simple explanation is that they are words that carry more visual details within them for instance instead of simply saying cat we can say tabby cat or Alley Cat or Baron von cat each one of these different keywords carries a different meaning and gives different results mid-journey is so effective that you can simply input your subject act or not without any other components and get spectacular results spectacular but not consistent not control random this brings us to the next component styling keyword is primarily there to control the feelings evoked by the image it's kind of like The Vibes of an image adjectives like elegant or angry bright colorful or devilish let's let's take the example of a cat again we'll use the basic non-packed cat subject to better illustrate the point we can see that applying the style keywords refines our original output in a certain direction now this works the same with all of the different subjects and it can be fascinating to see how certain subjects and Styles look when paired you can really see that it adds character and Flair to the initial subject this is key if you're trying to achieve your own artistic vision and to maintain the consistency throughout multiple images necessary for a project like a children's book for example but there's more to Artistic Vision than just the style of an image or the subject itself there's also the medium you can think of the medium as the foundation of the image keywords like impressionist photorealistic concept art claymation or black and white photography just to name a few to this point all of our examples have had a medium included we just didn't get any control over that Medium to take control we need to input our own medium keywords as we can see same example of our cat medium keywords really change the overall look of the image and taking a look at the other examples we can really see how powerful these medium keywords can be so now we have a really solid prompt in the works and in most situations your prompt is good to go from here but there are additional modifications we can make this is where modifiers come in not all AI image generators have modifiers though many do mid-journey in particular has a robust slew of modifiers we can add to our prompts you can find these modifiers in the mid-journey docks there are a lot of options and frankly this subject deserves an entire video which I'll be creating soon so make sure to subscribe but at a surface level it can be important to note that modifiers while the least necessary of all the components are still really powerful in many circumstances using dash dash AR to get the right size for YouTube videos versus Instagram photos or if you're feeling stuck in achieving your intended results you can use S10 if you really want to a confined mid-journey to your keywords or s a thousand if you want to let mid Journey's creativity shine or as mid-journey itself describes low stylization values produce images that closely match the prompt but are less artistic High values create images that are very artistic but less connected to the prompt I've found that I use AR and S all the time now it's time for the comprehensive example I decided to use Dragon to start with because I haven't used Dragon yet thus far we have this great 4x4 of a dragon it looks cool it's great right off the bat but it's not there yet let's push and pull it with our four component structure so the first thing I wanted to do was add the modifier for the aspect ratio at the end I know it's the last component but I found that if you have an intended aspect ratio in mind just put it in there right off the bat because aspect ratio does kind of affect the look of an image so I did just that I added my favorite aspect ratio the 1920x1080 and it definitely changes the image a lot but this Dragon looks really mean so I wanted to change that make them a little bit nicer so here we're faced with the first opportunity to add a new component and let's do so let's add the style component I'm going to add just one keyword flowery and you can see from the output of this image it looks a little bit nicer it still kind of looks like a very intense Dragon so I'm gonna push and pull it a little more before we get into adding the medium component I like to really Zone in on my exact subjects and styles so let's do that let's refine our subject and style further I think this Dragon still needs to be a little cuter so let's add tiny and cute so you can see from this output it totally changed it now we're talking this is a direction that I could get behind the problem here in my opinion is a lot of these kind of look toy-like so let's refine further by adding our first medium keyword because now we're actually talking about changing the overall look of the image more so than just a style can do so I added the photorealistic medium keyword and it does change it the bottom right especially looks like a real dragon thing it definitely looks less like a toy so then I thought mid-journey seemed to want to make these into little cute toy looking thing so what if I just work with it what if I just add the medium keyword claymation instead of the other ones leaving all else the same and we get this really cute clay action figury looking creature and I like this but this wasn't what I was going for I wanted something that looks fantasy and magical I wanted something that didn't just look like a toy that you lost in the woods here I'm kind of beating around the bush because I've been using mid-journey a while and there's a couple of keywords that I just gravitate towards concept art being one of and let's apply it now this image is exactly what my original Vision was it's a dragon creature who's not mean he's cute and tiny but he doesn't look like a toy it looks like something you can see in a children's book an illustration something you could see in a video game I love this bottom right one best I'm also a huge fan of the color scheme I never actually said cotton candy Dragon but he does look like a little cotton candy Dragon so this is my favorite image thus far and I upscale it whoa hang on there little guy bringing images to life with animation is for another tutorial cool though isn't it so when we look at the journey from our first to last image we can see how powerful the four component prompt structure can really be okay you now have the exact prompt structure to help you achieve your Perfect Image there's a lot more to prompting for AI images than this but this should give you a good foundation for your prompt engineering journey I really hope you enjoyed this video and found it useful leave a like to show your support and subscribe to keep up to date with mid journey and all things AI

Channel: The Prince of Prompting

Views: 17,752

Keywords: Prompt Engineering, Midjourney, ChatGpt, Ai, Artificial Intelligence, Futurism, Future Tech, Mastering the Perfect Midjourney Prompt Formula, Every Midjourney Prompt Should Use This Formula, Midjourney perfect prompt formula, Midjourney prompts, Midjourney tutorial, artificial intelligence, prompt engineering, how to use Midjourney, Midjourney examples, Midjourney explained, best Midjourney prompts, Midjourney prompt, Midjourney AI, prompt engineer, runway ai, leonardo ai

Id: DaCzczC-1hY

Length: 8min 46sec (526 seconds)

Published: Sat Sep 16 2023