Join our AI Art Community on YouTube
YouTube

How to Create a Midjourney Comic Book

Are you planning to create your very own comic book?

However, despite how easy it is to create stunning imagery, maintaining a more or less consistent look throughout the comic is still a challenge.

In this guide, I’ll show you how I created the short graphic novel “Treasures Beyond Gold”.

Let’s dive right in.

Blog Minibanner
Like this content? Then share it!

How to Create a Midjourney Comic Book

I created a Midjourney comic book (sometimes also called a graphic novel) using nothing but ChatGPT, Midjourney, and the Affinity Suite.

And so can you!

Meet Maya, the female hero of our story, and Leo, her male partner in crime.

In today’s guide, I’ll show you how you can create an entire storyline, character concepts, individual scenes, and even the necessary dialogue for a full-blown graphic novel or comic book.

And guess what?

So can you and I’ll show you exactly how it’s done.

Whether you prefer to watch the full YouTube video or just want to read through all the steps right here, is entirely up to you.

Key Points (tl;dr)
  • ChatGPT and Midjourney are massive enables for creators who are able to connect the dots and harness the power of these tools to build products that people love.
  • The bigger challenge is ensuring a more or less consistent art style and look of your characters throughout the entire Midjourney comic book.
  • Remember that some scenes may require far more than just a single image panel in order to convey what is actually happening in the story.

The Story

Infographic showing how I created the story for my Midjourney comic book with ChatGPT
Infographic showing how I created the story for my Midjourney comic book with ChatGPT

So first of all, I start off by opening up ChatGPT.

I then start a new chat thread and switch to GPT-4.

I’m using GPT-4 simply because I’ve found it to be far more reliable in terms of following my exact instructions.

GPT-3.5 sometimes has a mind of its own and no matter how much I insist on it doing something specific, it will often ignore my request.

GPT-4 also seems to be much better at understanding the entire chat history and responding to contextual questions.

So at this stage, I usually start with a foundational prompt that sets the scene.

This prompt is much longer than any of my follow-up questions during the conversation.

For this particular exercise, my foundational prompt consists of 4 separate parts.

You are also an excellent storyteller and graphic novel author. Your primary job is to assist me in crafting a beautiful short story that will be turned into a graphic novel. The graphic novel will contain panels created with Midjourney.

This primes ChatGPT for everything that follows and it will take on the persona of a storyteller.

You can see that I used the word “graphic novel” here but for the purposes of our Midjourney comic book, it doesn’t really matter that there is actually a difference.

This part is quite important because it influences how ChatGPT will respond to my follow-up questions as well.

Here’s the second part of the prompt:

INPUT VARIABLES:
{genre} = Adventure
{theme} = Two treasure hunters searching for a long lost historical treasure in Southeast Asia
{pages} = 4

You might be wondering why I’m defining variables in my prompt.

Well, first of all, it makes it much easier for me to re-use this prompt at a later stage and simply change specific elements of the prompts.

I don’t have to search through the entire prompt and look for all the different references.

And I can easily re-use these variables in later sections of the prompts as well, just in case I want to expand the story of my Midjourney comic book.

So here I define the genre of our story as well as the broad theme.

I also defined a specific number of pages which was meant to guide ChatGPT in terms of the length of the story.

However, I must admit that I messed up this part because parges aren’t really a good measure in this particular case.

Let’s move on to part 3 of the prompt:

STORY:
Please come up with a short story based on {theme} that uses elements that are typical for {genre} stories. Provide me with a 3-4 sentence summary and store it in {plot}.

In this part tell ChatGPT that I want it to come up with a story for my Midjourney comic book.

And as you can see, instead of explicitly entering my genre or theme, I’m referencing the variables instead.

I’ve also asked it to store the result in a new variable, just in case I need it later on.

And now let’s move on to the last part of the prompt.

SCENES:
Then break down the story into a reasonable amount of scenes that would fit onto {pages} pages of a graphic novel. Describe each scene in 20 words or less. Where suitable, please add brief dialogue that captures the essence of the scene. We will use this dialogue for speech bubbles in the graphic novel.

This is the part where I ask ChatGPT to break the story down into scenes that fit into a certain number of pages.

I also ask it to add some dialogue as well.

Technically, you could also turn the word count into an input variable as well.

This prompt is merely a demonstration of what I did for this exercise.

So let’s have a look at what ChatGPT came up with.

It gave me a lengthy response, including several scenes but let’s look at the plot first.

{plot} = Two rival treasure hunters, Maya and Leo, begrudgingly team up to find a long-lost historical treasure hidden in Southeast Asia. As they face dangerous challenges together, they eventually form an unbreakable bond and discover the true meaning of partnership.

ChatGPT

Sounds like your typical treasure hunter adventure story and this doesn’t really tell us all that much about our Midjourney comic book.

But if we take a closer look at the scenes that ChatGPT has created, then we get a pretty good idea of what the full story looks like.

So let’s do that.

Scene Crafting

Infographic showing how ChatGPT also crafted all the scenes for my Midjourney comic book.
Infographic showing how ChatGPT also crafted all the scenes for my Midjourney comic book.

Here’s the part of the ChatGPT response that I didn’t show you earlier.

As you can see, it produced a total of 7 different scenes, each with a brief description and some basic dialogue.

Based on this, our current story of our Midjourney comic book goes a bit like this:

  1. Maya and Leo accidentally bump into each other on a Southeast Asian market while each one is staring at a map.
  2. They reluctantly agree to work together and try to decipher the map.
  3. Once they do, they take off into the jungle and encounter a river crossing, which they successfully overcome with teamwork.
  4. They find a hidden temple, solve some puzzles and gain access to a treasure chamber.
  5. They find the treasure but face the dilemma of having to choose greed or friendship.
  6. They choose friendship and walk out of the temple together.
  7. The story closes with both of them standing on a cliff, looking out at the landscape.

Ok, so this is already pretty good, however, I’m not sure whether these are enough scenes for our Midjourney comic book.

You probably noticed how there are some significant gaps between each scene.

It’s not entirely clear how they get from one scene to the next.

So I gave ChatGPT some feedback and asked it to elaborate.

That's not bad, but I don't think those are enough scenes. We need to fill in the gaps or add detail. Using the same story, create a total of 12 scenes with dialogue.

And it did just that.

It added considerably more detail and even though there still are some gaps, we can potentially fill them ourselves.

Or we can treat them like the end of chapters.

Remember, I’m not trying to create a full-length graphic novel, I’m just trying to show what the process might look like.

Ok, so now that we have our story outline, we can probably start creating the images for our Midjourney comic book, right?

Technically, yes, but we still don’t really know what our characters look like.

So let’s change that.

The Characters

Infographic showing the characters created with ChatGPT and Midjourney
Infographic showing the characters created with ChatGPT and Midjourney

So we already know that our 2 main characters are Maya and Leo but we know very little about them.

But of course, that’s not a problem, because ChatGPT will happily oblige and share everything we need to know.

So I started off by asking it to describe them.

Please describe each character of the story in 20 words or less.

And here’s what ChatGPT responded:

Maya:
Resourceful, determined, and fiercely independent, Maya is a seasoned treasure hunter with a sharp wit and a kind heart.

Leo:
Adventurous, charismatic, and intelligent, Leo is a skilled explorer who values loyalty and friendship above all else.”

ChatGPT

While this says a lot about our characters in terms of their personality and strengths, it’s not exactly what we need at this stage.

Sure, these are important elements of the story but ChatGPT has already integrated this knowledge.

Instead, we want to know what they look like.

So I tried a different type of prompt.

I actually wanted you to describe what each character looks like, how old they are, and what sort of clothes they are wearing.

And this time ChatGPT understood what I wanted to hear from it and produced two very brief but good descriptions of Maya and Leo for our Midjourney comic book.

Maya:
Late 20s, athletic build, shoulder-length dark hair. Wears a green cargo vest, khaki shorts, and sturdy boots. Carries a satchel.

Leo:
Early 30s, muscular, short-cropped brown hair, stubble. Wears a worn-out red bandana, beige shirt, cargo pants, and hiking shoes.”

ChatGPT

You could probably input these directly into Midjourney and get a decent set of images.

However, I usually take these elements and use them to craft an entirely new prompt.

We also haven’t defined any art style for our Midjourney comic book yet.

So let’s do that and create portrait images of Maya and Leo.

The Art Style

Infographic showing examples of the art style we used in my Midjourney comic book.
Infographic showing examples of the art style we used in my Midjourney comic book.

Everyone has different tastes in style.

And depending on the type of graphic novel you’re working on, you’ll have to experiment with different prompt elements until you get it right for your Midjourney comic book.

I personally like to use combinations of different art styles because there’s rarely one specific term that encapsulates exactly what I want.

For this graphic novel, I’ve decided to use a prompt that looks like this:

Blend of comic book art and lineart in full natural colors.

This combines elements of classic comic book art with the high level of detail that comes with lineart.

The reason I emphasize “full natural colors” is because Midjourney often associates the concept of lineart with black-and-white imagery.

I will usually place this prompt at the beginning of every single scene that I create.

Why the beginning?

Well, in my experience, especially when working with much longer prompts in Midjourney v5, defining the style at the beginning of the prompt has been very effective.

For this exercise, I’m not fixated on keeping the characters 100% consistent.

My objective is to keep the style roughly the same.

So let’s have a look at our prompt for Maya.

/imagine
Blend of comic book art and lineart in full natural colors. A full body portrait shot of an attractive Asian woman in her late 20s, with an athletic build and shoulder-length dark hair, wearing a green cargo vest and kakhi shorts with a sturdy boots, carrying a satchel, tropical rainforest in the background --v 5 --ar 4:5 --s 750"

As you can see, I’ve incorporated most of the features of Maya that ChatGPT gave me.

I’ve set the aspect ratio to 4:5 because I’m only creating a sample portrait and not our Midjourney comic book.

And I’ve set the stylize parameter to 750 because I want it to return more creative images but nothing too crazy.

We’ll mostly keep this value throughout the entire process.

And in the image at the top of this section, you can see what Maya roughly looks like according to Midjourney.

Remember, I said I’m not going to be overly strict with regard to our character consistency.

Today’s guide is exclusively about the process of creating a Midjourney comic book, not character consistency.

I’m not interested in hearing people lament about how all these tools aren’t able to produce exactly what you want.

If you’re that picky, go learn how to draw.

Problem solved.

So next, let’s check out what Leo’s prompt looks like.

/imagine
Blend of comic book art and lineart in full natural colors. A full body portrait shot of an attractive and muscular man in his early 30s, with short-cropped brown hair and a stubble beard, wearing a bandana and a shirt in beige color and cargo pants and hiking shoes, tropical rainforest in the background --v 5 --ar 4:5 --s 750

As you can see, the logic is mostly the same.

I’ve only replaced the character’s features.

Here’s what Leo’s portraits look like.

Once again, these aren’t 100% consistent and that’s fine for our Midjourney comic book.

They’re good enough for what we’re trying to do.

These images might seem familiar to you because I shared these earlier this week on social media.

If you’d like to get more real-time updates and insights into what I’m working on, then follow me on Twitter.

I share new images, ideas, and prompts nearly daily and it’s a great way to stay in touch too.

So now that we picked a style that works for us, it’s time to start working on the images for the various different scenes.

The Panels

Ok, so now is when the real fun starts.

However, it also happens to be one of the more challenging aspects of creating a Midjourney comic book.

Asking ChatGPT to come up with a story and scenes is one thing.

But even though Midjourney will help you create the images, you’re going to have to put some thought into the sizes and shapes of your panels.

Some scenes will require more than one image, especially if you can’t fit all of the action into a single one.

ChatGPT could probably help you with this as well, but I decided to figure this out myself.

I mean, what’s the point if there isn’t at least some creative element to this process?

Before I start, I’d just like to give you some fair warning.

I’m not going to be able to explain every single detail on every prompt and image.

There’s quite a bit of material to go through and I really don’t want to bore you too much.

So let’s get started with Scene Number 1 of our Midjourney comic book.

Scene 1
Infographic showing the panels for Scene 1 of the Midjourney comic book.
Infographic showing the panels for Scene 1 of the Midjourney comic book.

In a bustling Southeast Asian market, Maya and Leo accidentally bump into each other while following the same ancient map.

Dialogue:
Maya: “Hey, watch where you’re going!”
Leo: “That’s my line!”

ChatGPT

Here’s the prompt I used for this image.

/imagine
Blend of comic book art and lineart in full natural colors. Attractive Western man in his early 30s with short-cropped brown hair and stubble beard and shirt in beige color walking through bustling Southeast Asian market, reading a treasure map --v 5 --ar 9:8 --s 750

Let me quickly explain the structure here a bit.

I start with the style prompt.

I then follow with the prompt that describes my character.

And then I follow up with activities and locations.

You don’t have to do it this way and you’ll see me switch things around later on, but usually works quite well.

You probably also noticed the aspect ratio of 9:8.

I chose this because I’m going to use smaller panels for this.

You’ll see what I mean at the very end.

Here’s what we get from this prompt.

Next up, let’s create something similar for Maya.

/imagine
Blend of comic book art and lineart in full natural colors. Attractive Asian woman in her late 20s with shoulder-length dark hair, wearing a green cargo vest walking through bustling Southeast Asian market, reading a treasure map --v 5 --ar 9:8 --s 750

The mostly the exact same prompt, except that I’ve replaced the character’s features.

And this gives us the following image.

So basically the very same type of scene but with Maya in it.

I still need a 3rd image though, because we need to convey the fact that they’ve bumped into each other.

This is what I meant when I said that you’ll need more than one image for each scene of your Midjourney comic book.

/imagine
Blend of comic book art and lineart in full natural colors. In a bustling Southeast Asian market, an attractive Asian woman in her late 20s with shoulder-length dark hair wearing a green cargo vest and an attractive man in his early 30s with short-cropped brown hair and a stubble beard wearing a shirt in beige color accidentally bump into each other while following the same ancient map. --v 5 --ar 9:4 --s 750"

In this prompt, you’ll notice that I start with the style but then I immediately set the scene first.

Only then do I add the characters and finally end with the activities.

As I said, you will need to experiment with what works best for you.

I’ve also changed the aspect ratio because I’m creating a wider image panel.

Ok, so this completes our first scene.

Whether you use just one or multiple images is really up to you.

In today’s exercise, you’ll see that sometimes 1 is enough, while in other cases I have to use more than 1 image for our Midjourney comic book.

And yes, I am absolutely aware of the fact that the faces and clothes are not consistent throughout all of these.

As I’ve already said, that’s not the point of today’s guide.

You can make these far more consistent with some simple “photobashing” and image prompts.

But the process is too time-consuming for this guide.

Let’s move on to Scene number 2.

Scene 2
Infographic showing the panels for Scene 2 of the Midjourney comic book.
Infographic showing the panels for Scene 2 of the Midjourney comic book.

Maya and Leo reluctantly agree to work together, deciphering the map’s cryptic riddles.

Dialogue:
Maya: “Fine, but we split the treasure!”
Leo: “Deal.”

ChatGPT

Here’s the prompt I used for the first image.

/imagine
Blend of comic book art and lineart in full natural colors. A Western man with stubble and an Asian woman deciphering riddles on a map. Interior architecture of a bar in the background. --v 5 --ar 9:4 --s 750

This prompt is much shorter a focused on the surroundings because I need something that conveys the activity in our Midjourney comic book.

The resulting image isn’t perfect, but it does a good job of showing what’s going on in the scene.

I need one more image though.

/imagine
Blend of comic book art and lineart in full natural colors. Inside of a tropical bar. An attractive Asian woman in her late 20s with shoulder-length dark hair wearing a green cargo vest and an attractive Western man in his early 30s with short-cropped brown hair and a stubble beard wearing a shirt in beige color deciphering riddles on a map. --v 5 --ar 9:4 --s 750

Back to the core prompt, except that we’re not in a tropical bar and both are deciphering riddles on a map.

Cool, so we now have all our images for the second scene.

Let’s move on to Scene number 3.

Scene 3
Infographic showing the panels for Scene 3 of the Midjourney comic book.
Infographic showing the panels for Scene 3 of the Midjourney comic book.

The duo embarks on their journey, navigating through a vibrant village filled with locals.

Dialogue:
Maya: “We should ask around for clues.”
Leo: “Good idea.”

ChatGPT

For this one, we only need a single image.

/imagine
blend of comic book art and lineart in full natural colors. In a village full of indigenous people, an attractive Asian woman in her late 20s with shoulder-length dark hair wearing a green cargo vest and an attractive man in his early 30s with short-cropped brown hair and a stubble beard wearing a shirt in beige color embark on a journey in search of treasure. --v 5 --ar 9:8 --s 750

This time I’m going to use an aspect ratio of 9:8 again because I want to create a bigger panel for our Midjourney comic book.

And here’s the image of Maya and Leo walking the village.

This transitions us straight into Scene number 4.

Scene 4
Infographic showing the panels for Scene 4 of the Midjourney comic book.
Infographic showing the panels for Scene 4 of the Midjourney comic book.

Maya and Leo receive vital information from an elderly villager about the treasure’s history.

Dialogue:
Elder: “The treasure is cursed; it tests the hearts of those who seek it.”

ChatGPT

Once again, a single image should be sufficient for this.

/imagine
blend of comic book art and lineart in full natural colors. An Asian woman in her late 20s with shoulder-length dark hair wearing a green cargo vest beings a conversation with an extremely old indigenous man in a tropical village. --v 5 --ar 9:4 --s 750

Since placing 3 characters into a scene is always a bit tricky in a Midjourney comic book, I’ve opted to leave out Leo.

Only using Maya makes this much easier to do.

And I think this one turned out quite alright.

Moving on the Scene number 5.

Scene 5
Infographic showing the panels for Scene 5 of the Midjourney comic book.
Infographic showing the panels for Scene 5 of the Midjourney comic book.

Journeying through the dense jungle, the duo encounters a treacherous river crossing, testing their teamwork skills.

Dialogue:
Maya: “We’ll need to build a raft.”
Leo: “On it.”

ChatGPT

This scene is a bit trickier because it requires close-ups as well as zoomed-out angles.

We’ll probably need 3 images for the whole thing.

Here’s the first one.

/imagine
blend of comic book art and lineart in full natural colors. In tropical jungle, an attractive Asian woman in her late 20s with shoulder-length dark hair wearing a green cargo vest and an attractive man in his early 30s with short-cropped brown hair and a stubble beard wearing a shirt in beige color attempt a river crossing. --v 5 --ar 9:4 --s 750

I need an image showing both our characters walking through the jungle and a river should be close by.

I think this one does a pretty good job.

However, I also need an image showing how they arrive at the shoreline of the river.

/imagine
blend of comic book art and lineart in full natural colors. River crossing inside a tropical jungle. A male and a female explorer stand on shore of a large river. --v 5 --ar 9:4 --s 750

For this one, I’ve opted to leave out our character descriptions again because we’re focusing on the surroundings.

And as you can see, this works really well because the characters are so small anyway.

But we still need one final image to close out the scene.

/imagineĀ 
Blend of comic book art and lineart in full natural colors. A Western man with stubble and an Asian woman building a river raft in the jungle. --v 5 --ar 9:8 --s 750

Once again, I’ve chosen a much shorter prompt for this scene of our Midjourney comic book.

And even though the consistency of the characters suffers a lot, it does a great job of conveying the activity.

Let’s move on to Scene Number 6.

Scene 6
Infographic showing the panels for Scene 6 of the Midjourney comic book.
Infographic showing the panels for Scene 6 of the Midjourney comic book.

While resting at their campsite, Maya and Leo share stories about their past adventures, finding common ground.

Dialogue:
Maya: “I never expected to meet someone who understands this life.”
Leo: “Me neither.”

ChatGPT

For this scene of our Midjourney comic book, I’m going to need a total of 2 images.

Luckily, I can probably squeeze out 2 images with the same prompt.

/imagine
blend of comic book art and lineart in full natural colors. Inside a jungle, an attractive Asian woman in her late 20s with shoulder-length dark hair wearing a green cargo vest and an attractive man in his early 30s with short-cropped brown hair and a stubble beard wearing a shirt in beige color sitting around a campfire. --v 5 --ar 9:4 --s 750

And indeed, by simply re-rolling the prompt a few times, I was able to extract these two images.

One simply shows Maya and Leo at their campfire site.

And on the other one Maya actually has a slightly concerned look on her face, which is perfect.

Because in scene 7, things get dangerous.

Scene 7
Infographic showing the panels for Scene 7 of the Midjourney comic book.
Infographic showing the panels for Scene 7 of the Midjourney comic book.

Facing a dangerous snake, the pair works together to avoid a deadly bite.

Dialogue:
Maya: “Distract it, I’ll grab a stick!”
Leo: “Got it!”

ChatGPT

For this scene, here’s the prompt I used.

/imagine
Blend of comic book art and lineart in full natural colors. During night time, a western man and an asian woman fight together against a snake in the deep jungle. --v 5 --ar 9:4 --s 750

And once again, I was able to extract two images from this.

One image in which both are approaching the snake.

And another one where Maya is fighting the snake.

Alright, let’s move on to Scene number 8.

Scene 8
Infographic showing the panels for Scene 8 of the Midjourney comic book.
Infographic showing the panels for Scene 8 of the Midjourney comic book.

The duo discovers a hidden temple entrance and solves a puzzle, revealing the entrance to the treasure chamber.

Dialogue:
Maya: “We did it, Leo!”
Leo: “Together, we’re unstoppable!”

ChatGPT

For this scene, we need at least 3 images, but I’ll try to go for 4.

Here’s the first prompt.

/imagine
Blend of comic book art and lineart in full natural colors. During the night, a Western man with stubble and an Asian woman standing before a ancient temple entrance covered in runes and foliage in the jungle. --v 5 --ar 9:4 --s 750

By simply re-rolling this prompt a few times, I was able to extract 2 suitable images for our Midjourney comic book already.

This image right here shows them approaching a temple-like structure.

And this image shows a different perspective as they walk toward an entrance.

Next, I need an image suggesting that they’re solving a riddle.

/imagine
blend of comic book art and lineart in full natural colors. Inside a dark ancient temple, an attractive man in his early 30s with short-cropped brown hair and a stubble beard wearing a shirt in beige color solving a riddle. --v 5 --ar 9:4 --s 750

This prompt should give me an image of Leo standing within the temple.

And sure enough, even though this really isn’t a perfect image, I guess this works.

I still need one more image that suggests that they gain access to the chamber.

/imagine
blend of comic book art and lineart in full natural colors. Inside a dark temple chamber made of stone, an attractive Asian woman in her late 20s with shoulder-length dark hair wearing a green cargo vest and an attractive man in his early 30s with short-cropped brown hair and a stubble beard wearing a shirt in beige color discover tombs full of gold. --v 5 --ar 9:4 --s 750

I had to re-roll this one quite a few times, but I finally settled on the following image for our Midjourney comic book.

I like this one because it shows that they’re happy to have found the treasure.

Now let’s move on to Scene number 9.

Scene 9
Infographic showing the panels for Scene 9 of the Midjourney comic book.
Infographic showing the panels for Scene 9 of the Midjourney comic book.

Inside the chamber, Maya and Leo find the treasure but also face a choice: greed or friendship.

Dialogue:
Maya: “We could just take it allā€¦”
Leo: “But at what cost?”

ChatGPT

Here’s the prompt.

/imagine
blend of comic book art and lineart in full natural colors. Ancient temple chamber. Dozens of wooden boxes full of gold bars and coins and other precious items. --v 5 --ar 9:8 --s 750

This one is quite easy because I can use the same prompt with 2 different aspect ratios to get 2 suitable images.

Here’s the image with an aspect ratio of 9:8.

And here’s another variation using an aspect ratio of 9:4.

We’re getting closer to the finish line now. Here’s scene number 10.

Scene 10
Infographic showing the panels for Scene 10 of the Midjourney comic book.
Infographic showing the panels for Scene 10 of the Midjourney comic book.

Choosing friendship over greed, the two leave the treasure behind, triggering a trap that seals the chamber.

Dialogue:
Maya: “That was a close one!”
Leo: “We’re better off without that curse.”

ChatGPT

So for this scene of our Midjourney comic book, I’ll probably need 3 images again.

But instead of showing you every single prompt, I’m going to finish off these last few scenes by only showing you the images I picked.

Remember, the process is largely the same for every single one of these.

This image shows how Maya and Leo have their doubts.

In the meantime, the sun has come up over the jungle.

They decide to walk away instead.

Exiting the chamber triggers a trap that seals the door.

Maya turns around one last time.

And then Scene 11 begins.

Scene 11
Infographic showing the panels for Scenes 11 & 12 of the Midjourney comic book.
Infographic showing the panels for Scenes 11 & 12 of the Midjourney comic book.

Maya and Leo stand side by side on a cliff, looking out at the vibrant Southeast Asian landscape, celebrating their new bond.

Dialogue:
Maya: “We found something more valuable today.”
Leo: “Friendship.”

ChatGPT

I picked this image because I wanted something big enough to cover 2/3 of the page.

And I’ll use this one for the dialogue.

And finally, the story ends with scene number 12.

Scene 12

Ready for their next adventure, the duo walks into the sunset, excited for the challenges ahead.

Dialogue:
Maya: “Where to next, partner?”
Leo: “The world is ours to explore!”

ChatGPT

Here I do more or less the exact same thing, except that I’ll start with a wider image first.

Followed by a nice big one to cover the rest of the page.

And that concludes the full story of our graphic novel.

But we’re still missing one essential part.

We need a title and a cover image.

So let’s get to work!

The Cover

Infographic showing how we created the cover image for the Midjourney comic book.
Infographic showing how we created the cover image for the Midjourney comic book.

I asked ChatGPT for some suggestions on what to call our Midjourney comic book.

Here’s how I asked it:

Come up with 3 suggestions for a title for the graphic novel.
Please also describe in 20 words or less what the cover image looks like.

It then suggested these 3 titles.

And it also provided a description for a cover image.

Cover Image:
Maya and Leo, back-to-back, wielding their tools, with a dense jungle, hidden temple, and ancient map in the background.

ChatGPT

Using this description as a template, I built the following prompt for Midjourney.

/imagine
Blend of comic book art and lineart in full natural colors, An attractive western man with stubble and an attractive Asian woman with shoulder-length dark hair, back-to-back, wielding their tools, with a dense jungle, hidden temple, and ancient map in the background:: 
adventure movie poster, western man and asian woman:: 
text, title, headline, words::-0.5 
--v 5 --ar 4:5 --s 750

You probably noticed that I’m using a multiprompt here.

The second segment is meant to apply more of a movie poster look to the cover of our Midjourney comic book.

While the third segment ensures that we get as little text as possible.

Now check out the image we get (see the image above).

I agree that it might look a bit too much like a movie poster, but I really like it.

And if I add “Treasures Beyond Gold” as a title, then it looks even better.

Now all I needed to do was put it all together into a nice graphic novel layout using Affinity Publisher.

And the overview of all the spreads looks like this.

Since I know how much you guys love to work with templates, I’ve put together a folder on Google Drive.

  • It includes all the ChatGPT prompts and answers.
  • It also includes the Affinity Publisher template and the cover image.
  • It even includes every single image that I used in the graphic novel.

If you’d like to learn more about character design, storytelling, product design, or just plain old prompting in Midjourney, then check out some of my other guides on Midjourney or have a look at my YouTube channel.

This channel is meant to be a place for you to learn and improve your skills.

So keep on learning and take care!

Frequently Asked Questions (FAQ)

Before we close off this guide, let’s quickly address some of the most common questions related to Midjourney comic books.

  • Can I use AI to create a comic book?

    Yes, it is possible to create a comic entirely from AI tools. You can use ChatGPT to help write your storyline, describe scenes, write dialogue, and flesh out your characters. You can then use Midjourney to create all the visuals that you need for your Midjourney comic book.

  • Can Midjourney create comic books?

    Midjourney is an AI art generator first and foremost. It can be used to create comic book artwork, however, keeping the visuals consistent throughout the entire book is still a challenge.

  • How do I create a consistent character in Midjourney?

    Creating a consistent character is always a challenging process, even with Midjourney. There are certain techniques that allow you to achieve relatively consistent results, however, new obstacles arise as soon as you want to introduce more than one character. Check out this article for more details.

Conclusion

Creating a full-length Midjourney comic book or graphic novel is a challenging project that should not be taken lightly.

Despite all the help you can get from ChatGPT and Midjourney, you will undoubtedly hit certain obstacles along the way and you will also be surprised how many images you need to create in order to tell your story properly.

That being said, these AI tools have become massive enablers for creators around the world who are keen to produce content that others can enjoy.

Here at Tokenized, we want to help you learn as much as possible about the AI software industry. We help you navigate the world of tech and the digitalization of our society at large, including the tokenization of assets and services.

Total
0
Shares
Share
Tweet
Share
Share
Share
Pin it