Everything you need to know to MASTER MidJourney and create stunning pieces of art!
In this ultimate guide we're going to go though everything you need to know about Midjourney. The goal is that when you get to the end of this article you'll have a deep and full understanding of how to use this fantastic new tool. Also, we want this to be a valuable resource that you can always reference back to. So we're going to be keeping this article up to date with new Midjourney features and content.
There's a lot to cover so if you want to skip sections or jump back to where you left off, here are the topics we'll be covering:
Have you ever found turning your creative ideas into visually appealing designs challenging? Do you wish there was a simpler way to express your imagination and bring it to life? If so, you may want to learn more about Midjourney. Midjourney is a cutting-edge text-to-image AI model developed by a team of researchers and designers dedicated to expanding the limits of human imagination.
Similar to other AI programs like DALL-E and Stable Diffusion, Midjourney allows you to generate stunning images based on a text description using simple commands on Discord. Led by David Holz, co-founder of Leap Motion, the Midjourney team has created an innovative tool that has been embraced by artists and advertisers alike for its ability to rapidly prototype and brainstorm original concepts.
But Midjourney isn't just another AI program - it's the product of an independent research lab focused on exploring new mediums of thought and expanding human creativity. With 11 full-time staff (as of 02/2023) and a team of expert advisors, Midjourney is at the forefront of a new era of creative expression.
Ready to take your imagination to the next level? Keep reading to discover how Midjourney can help you unleash your inner artist.
There's a lot of excitement in the world of artificial intelligence and machine learning, with a few text-to-image models that are turning heads. Among them are Dall-E 2 and Stable Diffusion, which are getting some serious buzz. Dall-E 2 has been especially impressive, creating incredibly detailed images, like portraits and scenes. It's no wonder it's received so much recognition!
On top of that, Stable Diffusion has also made a big splash by making their model open-source, which means anyone can use it. How cool is that?
And then there is MidJourney, another text-to-image model that's been getting a LOT of attention, and some people even consider it the best of the bunch. Don't believe me? Just take a look at these amazing results!
You can achieve astonishing outcomes with the right prompt!
First, let's go through step-by-step how to get Midjourney up and running.
Your first step to unleashing your inner artist is to get a Discord login. If you don't have one, you can learn how to create one on the Discord website. After that, go to Midjourney.com and smash "Join the Beta," in the bottom right corner. Or go directly to the Midjourney Discord.
After you accept the invitation it should look something like this:
Once you're in the Midjourney Discord server, select any newbies-# channel visible on the left sidebar.
I know I'm excited too!
Now it'll take about a minute to generate a grid of 4 images from your prompt. When you sent your prompt, Midjourney added it to the queue to be generated when the server resources become available. As a trial user, you can make roughly 25 Jobs before needing to subscribe.
A Job is any action that uses the Midjourney Bot, including upscaling images, creating image variations, or using the /imagine command to create a grid of images.
Great! Now you should see a 2x2 grid of photos. Let's take a look at what we got from "Japanese Cherry Blossum" (Good thing AI understands that I meant to say "Blossom").
Pretty crazy that we can get these beautiful images from just a few minutes of work!
Now we have a few options, notice the buttons under the photos that look like this:
The U on the top row stands for Upscale
The V on the bottom row stands for Version
This little button 🔄 means Re-Roll. This returns the same job, so in this case it will make 4 new images from the prompt "Japanese Cherry Blossom".
Now, each photo is numbered 1-4 just like this:
If you like one or multiple of the images then just click the Upscale button with the number of the image you want upscaled. I love this first cherry blossom but I want to see what other versions it can generate. So I'll click "V1", this basically is saying "Give me 4 more Versions of image 1".
Now we have 4 slightly different versions of our 1st image. I just clicked on it to get a better view.
After checking out our lovely variations I like the 1st one the best again. So to make it larger we just click " U1" and we end up with this:
Perfect! From here you'll notice a few more buttons at the bottom, we'll cover these in a later section "Advanced Techniques - Preferences, Weights, and Image Prompts".
After upscaling an image, you can rate it by selecting one of the four smiley buttons on the bottom row or you can also rate it on the Midjourney website.
According to the Midjourney website "The top 1000 image raters each day receive an hour of free, Fast Mode time."
To save your image, simply click on it to open it to full size.
Then right-click and choose Save image. On mobile, long-tap the image and then tap the download icon in the top right corner. All images are also immediately available to view on Midjourney.com/app. You just have to sign in with Discord to view them.
When you sign-up you get 25 images free this includes prompt, variation, and upscale. If you want to create more images, then you have to sign up for a subscription plan.
To sign up you have to use the /subscribe command to generate a personal link to the Midjourney account page.
Don't share this personal link with others.
Here you can choose Monthly Billing:
Or you can save 20% by choosing Yearly Billing:
Most people will feel right at home using the Basic plan however there are some benefits worth mentioning to the other plans.
The most notable differences are:
Midjourney has two modes for image generation, “fast” and “relax”.
Fast tries to give you a GPU instantly.
Relax puts you in a queue behind others based on how much you've used the system in relax mode.
Now for the fun part! Here we'll walk through in-depth how to create and use your own Midjouney prompts.
We'll break this section into four parts:
A Basic Prompt only includes the description of the image you are wanting to create. This can vary widely depending on your desired image though as you'll soon see.
I've found there are 5 main points to consider when creating your basic prompt, these are:
We'll be structuring our prompt like this [ Style ] [ Subject ] [ Feeling ] [ Detail ] but don't worry, we'll build it block by block. Plus you can always write prompts as detailed or simple as you want.
The style of the prompt refers to the visual characteristics or features, such as the type of art, medium used, art style, artist style, or decade that the AI will consider while creating the image. For instance, if a prompt is designed with a watercolor and colorful style, the resulting image is likely to be vibrant and colorful as well. On the other hand, if the style is more subdued or muted, the image generated by the AI will reflect those characteristics.
We'll format it like this: [Style] of ~
If you would like to utilize Midjourney as a reference for Manga ideas, type in [image type] of ~ and whatever object you're wanting a style of.
Prompt example: Photograph of a castle
Other image styles to try:
You can also use different art mediums to generate some specific results!
Prompt : Watercolor Painting of a humming bird
Including an art style in your prompt is an excellent way to use a simple description and get some wildly different results.
Prompt example: Pixar art Fox
A few other art styles to try:
Generating an image in the style of your favorite artist can be a fun and exciting way to explore your creativity! To get started, simply include the name of the art style or title of the art piece you want to emulate. For instance, if you're looking to create an image inspired by Banksy's "Girl With Balloon", just type in "Girl With Balloon" along with "Banksy style". With these simple steps, you can unleash your imagination and create stunning artwork that captures the essence of your favorite artist!
Prompt example: Redwood tree Pablo Picasso style
And lastly, we can dramatically change our image by specifying a time period.
Prompt example: City street from the 1700s
A few other time periods to try:
After specifying the image style, the next step is to add a more descriptive touch - whether you're creating a person, location, or food photo, let your creativity take the lead and create a beautiful description that will be sure to wow!
Prompt Example: Red Mushroom Growing on a log in the rain
The focal point can be anything you want to create, from a mushroom to a mountain. Your imagination is the limit! Our goal here is to provide a precise description of what we want.
Here are a few examples to get your creative juices flowing:
Composition can heavily influence the viewer's perception of an image. Effective composition can make an image feel balanced, interesting, and visually striking, whereas poor composition can make it feel cluttered or uninteresting.
Prompt Example: Hyper-realistic Dragonfly, negative space composition
A few more composition ideas:
Also, don't forget to include the location your subject is in. Here we play with a few locations for a panda bear.
Prompt Example: Panda bear standing on an erupting volcano
Adding the correct emotional words is what really makes your image generations shine! You can use these to describe your subject or even your image as a whole. Let's take a look.
Prompt Example: Smiling African princess wearing a colorful clothes
Here are a few more emotional ideas for you to use:
Lighting can make a significant impact on the quality and atmosphere of an image. It can also help with giving depth to an image by exaggerating shadows and flattening features to highlight your image's best qualities.
Prompt Example: Moody Lighting - Tropical Italian coastal village on a cliff
Here are some more lighting ideas to try out:
We can also use specific colors or color tones to further refine the feeling we want to portray. check out some examples we made for a futuristic car!
Prompt Example: Monochromatic - Futuristic car driving on a busy city street at night
A few other examples of color:
Prompt Example: Saguaro cactus on a stary night
Aaand yet again, here are some more camera suggestions for you:
Prompt Example: Moody Tortoise with a bowler hat
3ds maxV-ray
Unreal engine
Octane render
8k,4kHQ
HD)
photo realistic
Midjurney commands are the core way to talk to the bot. We'll start by breaking down the most impressive and under the radar Midjourney commands then we'll go through the more basic ones!
Here we'll show you step-by-step how to leverage Midjourney creative commands to master your image generation and create original art that no-one even knew could be created.
In a nutshell, Multi-Image Prompting is a highly versatile tool that can blend images seamlessly. It could be its own application and I'd still be happy. No need to upload images and copy-paste URLs - with /blend you can simply drag and drop up to 5 images into their designated slots and hit enter.
You can also adjust the aspect ratio according to your preference. By default, the blend will be square, but you can explicitly set the aspect ratio using the Dimension parameter to avoid black stripes on the generation sides, especially when using a 2:3 picture.
This is the go-to parameter for creating shortcuts to your own created characters or scenery.
To use it simply
(More extensive guide coming soon)
This is how you manage your custom options.
View your current custom options.
Specify a suffix to add to the end of every prompt.
Toggle Remix mode.
Our more in-depth guide is coming soon, for now here is the Midjourney documentation: https://docs.midjourney.com/docs/remix
These are general commands to navigate through your discord.
This command allows you to generate an image using a prompt. Simply provide a prompt and the bot will create an image based on it.
If you're new to the Midjourney Bot, this command is a great place to start. It provides helpful information and tips about the bot, including how to use its various features and commands.
With this command, you can view information about your own profile, plan, usage, and more. This can be useful for keeping track of your activity on the platform and monitoring your usage.
This command generates a personal link for your account page. You can use this link to easily access your account and view your account details, plan information, and more.
This command switches the bot to fast mode. In this mode, the bot will complete jobs more quickly, but it will also cost money to use. This feature is only available to Pro Plan subscribers.
The relax mode is designed for users who don't want to spend money on generating images. Although generating images in this mode is free, it takes longer for the bot to complete jobs. This feature is only available for the Standard and Corporate plans.
Use this command to regenerate any job by providing its Job ID. This can be useful if you need to view a completed job or regenerate an image for further modifications.
By default, all jobs created with the Midjourney Bot are public. If you want to keep your jobs private, you can use this command to switch to private mode. This feature is available for an additional fee of $20 per month.
This command switches your jobs to public mode, meaning they can be seen by everyone else on the platform. This is the default mode for all jobs created with the bot, but you can switch to private mode if you prefer.
Midjourney parameters are inputs that you can add at the end of your command. By learning how to use these parameters, you can gain a lot of freedom and control in generating images with Midjourney.
With this command, you can decide how artistic you want your generation to be.
With this command, you can determine the time and quality of the generation of your image.
Adding a double colon (::) to a prompt in Midjourney Bot indicates that the prompt should be considered as separate parts. If a prompt is not separated, it is treated as a single thought. For instance, the prompt "hot dog" will produce images of tasty hotdogs, whereas "hot:: dog" will produce images of a dog that is warm.
Important Details to Note:
Examples:
Prompt Weights: Assigning Importance to Each Part of the Prompt
When using the double colon to separate a prompt, you can assign relative importance to each part of the prompt by adding a number immediately after the double colon.
Important Details to Note:
Examples:
Negative Prompt Weights: Removing Unwanted Elements
Negative weights can be added to prompts to remove unwanted elements. However, the sum of all weights must be a positive number.
Examples:
The --no Parameter: Same as Negative Prompt Weights
The "--no" parameter is the same as weighing part of a multi-prompt to "-.5". For example, "vibrant tulip fields:: red::-.5" is the same as "vibrant tulip fields --no red".
You now possess a greater understanding than roughly 7 billion people of how to construct Midjourney prompts. With this knowledge, you should feel confident in your ability to utilize Midjourney and create some truly remarkable images!
I'm proud of you for sticking through and I hope you've had some fun, Happy creating!
Everything you need to know to MASTER MidJourney and create stunning pieces of art!
In this ultimate guide we're going to go though everything you need to know about Midjourney. The goal is that when you get to the end of this article you'll have a deep and full understanding of how to use this fantastic new tool. Also, we want this to be a valuable resource that you can always reference back to. So we're going to be keeping this article up to date with new Midjourney features and content.
There's a lot to cover so if you want to skip sections or jump back to where you left off, here are the topics we'll be covering:
Have you ever found turning your creative ideas into visually appealing designs challenging? Do you wish there was a simpler way to express your imagination and bring it to life? If so, you may want to learn more about Midjourney. Midjourney is a cutting-edge text-to-image AI model developed by a team of researchers and designers dedicated to expanding the limits of human imagination.
Similar to other AI programs like DALL-E and Stable Diffusion, Midjourney allows you to generate stunning images based on a text description using simple commands on Discord. Led by David Holz, co-founder of Leap Motion, the Midjourney team has created an innovative tool that has been embraced by artists and advertisers alike for its ability to rapidly prototype and brainstorm original concepts.
But Midjourney isn't just another AI program - it's the product of an independent research lab focused on exploring new mediums of thought and expanding human creativity. With 11 full-time staff (as of 02/2023) and a team of expert advisors, Midjourney is at the forefront of a new era of creative expression.
Ready to take your imagination to the next level? Keep reading to discover how Midjourney can help you unleash your inner artist.
There's a lot of excitement in the world of artificial intelligence and machine learning, with a few text-to-image models that are turning heads. Among them are Dall-E 2 and Stable Diffusion, which are getting some serious buzz. Dall-E 2 has been especially impressive, creating incredibly detailed images, like portraits and scenes. It's no wonder it's received so much recognition!
On top of that, Stable Diffusion has also made a big splash by making their model open-source, which means anyone can use it. How cool is that?
And then there is MidJourney, another text-to-image model that's been getting a LOT of attention, and some people even consider it the best of the bunch. Don't believe me? Just take a look at these amazing results!
You can achieve astonishing outcomes with the right prompt!
First, let's go through step-by-step how to get Midjourney up and running.
Your first step to unleashing your inner artist is to get a Discord login. If you don't have one, you can learn how to create one on the Discord website. After that, go to Midjourney.com and smash "Join the Beta," in the bottom right corner. Or go directly to the Midjourney Discord.
After you accept the invitation it should look something like this:
Once you're in the Midjourney Discord server, select any newbies-# channel visible on the left sidebar.
I know I'm excited too!
Now it'll take about a minute to generate a grid of 4 images from your prompt. When you sent your prompt, Midjourney added it to the queue to be generated when the server resources become available. As a trial user, you can make roughly 25 Jobs before needing to subscribe.
A Job is any action that uses the Midjourney Bot, including upscaling images, creating image variations, or using the /imagine command to create a grid of images.
Great! Now you should see a 2x2 grid of photos. Let's take a look at what we got from "Japanese Cherry Blossum" (Good thing AI understands that I meant to say "Blossom").
Pretty crazy that we can get these beautiful images from just a few minutes of work!
Now we have a few options, notice the buttons under the photos that look like this:
The U on the top row stands for Upscale
The V on the bottom row stands for Version
This little button 🔄 means Re-Roll. This returns the same job, so in this case it will make 4 new images from the prompt "Japanese Cherry Blossom".
Now, each photo is numbered 1-4 just like this:
If you like one or multiple of the images then just click the Upscale button with the number of the image you want upscaled. I love this first cherry blossom but I want to see what other versions it can generate. So I'll click "V1", this basically is saying "Give me 4 more Versions of image 1".
Now we have 4 slightly different versions of our 1st image. I just clicked on it to get a better view.
After checking out our lovely variations I like the 1st one the best again. So to make it larger we just click " U1" and we end up with this:
Perfect! From here you'll notice a few more buttons at the bottom, we'll cover these in a later section "Advanced Techniques - Preferences, Weights, and Image Prompts".
After upscaling an image, you can rate it by selecting one of the four smiley buttons on the bottom row or you can also rate it on the Midjourney website.
According to the Midjourney website "The top 1000 image raters each day receive an hour of free, Fast Mode time."
To save your image, simply click on it to open it to full size.
Then right-click and choose Save image. On mobile, long-tap the image and then tap the download icon in the top right corner. All images are also immediately available to view on Midjourney.com/app. You just have to sign in with Discord to view them.
When you sign-up you get 25 images free this includes prompt, variation, and upscale. If you want to create more images, then you have to sign up for a subscription plan.
To sign up you have to use the /subscribe command to generate a personal link to the Midjourney account page.
Don't share this personal link with others.
Here you can choose Monthly Billing:
Or you can save 20% by choosing Yearly Billing:
Most people will feel right at home using the Basic plan however there are some benefits worth mentioning to the other plans.
The most notable differences are:
Midjourney has two modes for image generation, “fast” and “relax”.
Fast tries to give you a GPU instantly.
Relax puts you in a queue behind others based on how much you've used the system in relax mode.
Now for the fun part! Here we'll walk through in-depth how to create and use your own Midjouney prompts.
We'll break this section into four parts:
A Basic Prompt only includes the description of the image you are wanting to create. This can vary widely depending on your desired image though as you'll soon see.
I've found there are 5 main points to consider when creating your basic prompt, these are:
We'll be structuring our prompt like this [ Style ] [ Subject ] [ Feeling ] [ Detail ] but don't worry, we'll build it block by block. Plus you can always write prompts as detailed or simple as you want.
The style of the prompt refers to the visual characteristics or features, such as the type of art, medium used, art style, artist style, or decade that the AI will consider while creating the image. For instance, if a prompt is designed with a watercolor and colorful style, the resulting image is likely to be vibrant and colorful as well. On the other hand, if the style is more subdued or muted, the image generated by the AI will reflect those characteristics.
We'll format it like this: [Style] of ~
If you would like to utilize Midjourney as a reference for Manga ideas, type in [image type] of ~ and whatever object you're wanting a style of.
Prompt example: Photograph of a castle
Other image styles to try:
You can also use different art mediums to generate some specific results!
Prompt : Watercolor Painting of a humming bird
Including an art style in your prompt is an excellent way to use a simple description and get some wildly different results.
Prompt example: Pixar art Fox
A few other art styles to try:
Generating an image in the style of your favorite artist can be a fun and exciting way to explore your creativity! To get started, simply include the name of the art style or title of the art piece you want to emulate. For instance, if you're looking to create an image inspired by Banksy's "Girl With Balloon", just type in "Girl With Balloon" along with "Banksy style". With these simple steps, you can unleash your imagination and create stunning artwork that captures the essence of your favorite artist!
Prompt example: Redwood tree Pablo Picasso style
And lastly, we can dramatically change our image by specifying a time period.
Prompt example: City street from the 1700s
A few other time periods to try:
After specifying the image style, the next step is to add a more descriptive touch - whether you're creating a person, location, or food photo, let your creativity take the lead and create a beautiful description that will be sure to wow!
Prompt Example: Red Mushroom Growing on a log in the rain
The focal point can be anything you want to create, from a mushroom to a mountain. Your imagination is the limit! Our goal here is to provide a precise description of what we want.
Here are a few examples to get your creative juices flowing:
Composition can heavily influence the viewer's perception of an image. Effective composition can make an image feel balanced, interesting, and visually striking, whereas poor composition can make it feel cluttered or uninteresting.
Prompt Example: Hyper-realistic Dragonfly, negative space composition
A few more composition ideas:
Also, don't forget to include the location your subject is in. Here we play with a few locations for a panda bear.
Prompt Example: Panda bear standing on an erupting volcano
Adding the correct emotional words is what really makes your image generations shine! You can use these to describe your subject or even your image as a whole. Let's take a look.
Prompt Example: Smiling African princess wearing a colorful clothes
Here are a few more emotional ideas for you to use:
Lighting can make a significant impact on the quality and atmosphere of an image. It can also help with giving depth to an image by exaggerating shadows and flattening features to highlight your image's best qualities.
Prompt Example: Moody Lighting - Tropical Italian coastal village on a cliff
Here are some more lighting ideas to try out:
We can also use specific colors or color tones to further refine the feeling we want to portray. check out some examples we made for a futuristic car!
Prompt Example: Monochromatic - Futuristic car driving on a busy city street at night
A few other examples of color:
Prompt Example: Saguaro cactus on a stary night
Aaand yet again, here are some more camera suggestions for you:
Prompt Example: Moody Tortoise with a bowler hat
3ds maxV-ray
Unreal engine
Octane render
8k,4kHQ
HD)
photo realistic
Midjurney commands are the core way to talk to the bot. We'll start by breaking down the most impressive and under the radar Midjourney commands then we'll go through the more basic ones!
Here we'll show you step-by-step how to leverage Midjourney creative commands to master your image generation and create original art that no-one even knew could be created.
In a nutshell, Multi-Image Prompting is a highly versatile tool that can blend images seamlessly. It could be its own application and I'd still be happy. No need to upload images and copy-paste URLs - with /blend you can simply drag and drop up to 5 images into their designated slots and hit enter.
You can also adjust the aspect ratio according to your preference. By default, the blend will be square, but you can explicitly set the aspect ratio using the Dimension parameter to avoid black stripes on the generation sides, especially when using a 2:3 picture.
This is the go-to parameter for creating shortcuts to your own created characters or scenery.
To use it simply
(More extensive guide coming soon)
This is how you manage your custom options.
View your current custom options.
Specify a suffix to add to the end of every prompt.
Toggle Remix mode.
Our more in-depth guide is coming soon, for now here is the Midjourney documentation: https://docs.midjourney.com/docs/remix
These are general commands to navigate through your discord.
This command allows you to generate an image using a prompt. Simply provide a prompt and the bot will create an image based on it.
If you're new to the Midjourney Bot, this command is a great place to start. It provides helpful information and tips about the bot, including how to use its various features and commands.
With this command, you can view information about your own profile, plan, usage, and more. This can be useful for keeping track of your activity on the platform and monitoring your usage.
This command generates a personal link for your account page. You can use this link to easily access your account and view your account details, plan information, and more.
This command switches the bot to fast mode. In this mode, the bot will complete jobs more quickly, but it will also cost money to use. This feature is only available to Pro Plan subscribers.
The relax mode is designed for users who don't want to spend money on generating images. Although generating images in this mode is free, it takes longer for the bot to complete jobs. This feature is only available for the Standard and Corporate plans.
Use this command to regenerate any job by providing its Job ID. This can be useful if you need to view a completed job or regenerate an image for further modifications.
By default, all jobs created with the Midjourney Bot are public. If you want to keep your jobs private, you can use this command to switch to private mode. This feature is available for an additional fee of $20 per month.
This command switches your jobs to public mode, meaning they can be seen by everyone else on the platform. This is the default mode for all jobs created with the bot, but you can switch to private mode if you prefer.
Midjourney parameters are inputs that you can add at the end of your command. By learning how to use these parameters, you can gain a lot of freedom and control in generating images with Midjourney.
With this command, you can decide how artistic you want your generation to be.
With this command, you can determine the time and quality of the generation of your image.
Adding a double colon (::) to a prompt in Midjourney Bot indicates that the prompt should be considered as separate parts. If a prompt is not separated, it is treated as a single thought. For instance, the prompt "hot dog" will produce images of tasty hotdogs, whereas "hot:: dog" will produce images of a dog that is warm.
Important Details to Note:
Examples:
Prompt Weights: Assigning Importance to Each Part of the Prompt
When using the double colon to separate a prompt, you can assign relative importance to each part of the prompt by adding a number immediately after the double colon.
Important Details to Note:
Examples:
Negative Prompt Weights: Removing Unwanted Elements
Negative weights can be added to prompts to remove unwanted elements. However, the sum of all weights must be a positive number.
Examples:
The --no Parameter: Same as Negative Prompt Weights
The "--no" parameter is the same as weighing part of a multi-prompt to "-.5". For example, "vibrant tulip fields:: red::-.5" is the same as "vibrant tulip fields --no red".
You now possess a greater understanding than roughly 7 billion people of how to construct Midjourney prompts. With this knowledge, you should feel confident in your ability to utilize Midjourney and create some truly remarkable images!
I'm proud of you for sticking through and I hope you've had some fun, Happy creating!