MidJourney - More than just an AI image generator

[mostakimvip06]

MidJourney - what is it?
MidJourney is probably the largest and most popular generative AI application currently, which generates images from text phrases. Which is a unique phenomenon, because unlike the competing DALL-E, which is supported by the creator of the well-known Chat GPT - OpenAI, MidJourney is an independent and self-financed project.

To describe the entire process of MidJourney AI in detail, we would probably need at least a few dozen pages, so in today's post we will focus on the "cover", i.e. how to use this software, what its capabilities are, and why it is worth being interested in. However, if you would like us to create an entry describing the exact process of this platform, let us know in the comments!

Where did MidJourney come from?
MidJourney Inc. was founded in San Francisco, California by David Holz , who also co-founded the Leap Motion brand. The first beta of the MidJourney platform was launched on July 12, 2022, and the Discord server was created on March 14, 2022.

MidJourney has been working on improving its algorithms ever since, introducing new models every few months. This approach is appreciated by every modern digital marketing agency that focuses on innovative solutions. The current philippines telemarketing data model is V5.2, which went live on June 22, 2023. Below is a list of all published versions:

V1 - January 2022
V2 - April 2022
V3 - July 2022
V4 - November 2022
V5 - March 2023
V5.1 - May 2023
V5.2 - June 2023
MidJourney - how to install?
MidJourney is quite unique in terms of platform use compared to other AI tools, as it works through the popular communication application Discord.

By the way, if you are interested in useful AI tools that you absolutely must try, especially when working in the marketing and creative industry - check out our post about artificial intelligence (AI) in the work of a creative, marketing and interactive agency

Back to the topic, to get started with MidJourney:

1. We register in the Discord application
We can do this by going to the website and clicking the “Login” button. We fill out the form with account details and verify our email address by clicking the link we received.

Registration

2. For Windows, we recommend downloading the desktop app.
However, if you use other software, such as macOS, you can still use Discord in the browser app .

Registration process

After launching the Discord app, simply click the compass icon in the left panel and the MidJourney community should appear in the “featured communities.” However, if it is not visible in the featured communities, you can find it by typing MidJourney in the search bar.
Discord

We go to the “Getting started” section by clicking on this item
Getting started

At the top of the screen, click "Join MidJourney"
Joining MidJourney

We are in the MidJourney community! Now all we have to do is go to one of the “newbies” channels and we can get started!
Newbies

MidJourney - how to use?
To start generating images with MidJourney, we need to enter one of the “newbies” channels, the whole process is done in the text box at the bottom of the screen.

Newbies Channel

This is where we will enter all the prompts and generate our images.

What are the commands in MidJourney?
Depending on what we want to achieve, we can enter different commands in the text box, such as:

/imagine…
This is the most basic command we will use. It allows us to generate images, after typing “/imagine” and clicking space, an internal text field will appear where we describe what we would like to see.

Image generation

We confirm with the “enter” button. After a while, a message with our prompt and designation should appear (first it will be “waiting to start”, then the % progress will be shown)

Waiting to generate

After a while, MidJourney will generate 4 image suggestions based on our prompt.

Generated image suggestions

/info
The “/info” command will show us the details of our account, such as:

User ID
Active subscription type
Visibility mode
Remaining rednering time
Rednering time used so far
Renders queued
Active renders
/blend…
The “/blend” command allows us to combine two photos using AI.

For example, combining these two photos:

/blend command

This will give us a result something like this:

Connection result

/describe…
This command allows us to upload any graphic or photo, which MidJourney will then interpret for us and describe what is on it.

/help
It will present us with a panel with useful information and links. We encourage you to familiarize yourself with them.

/settings
It will open a panel with settings for generated images. Here we can specify options such as:

The MidJourney model version we want to use
Level of image stylization
Public/private mode
The degree of “variation” of generated images
Fast/slow mode
/prefer suffix
It allows us to automatically add options to generated images, which we discuss in more detail below.

How to detail prompts?
Of course, we can construct our prompts in a very simple way, i.e. by simply writing in one sequence the elements that we want to include in the generated graphic. However, if the result is not entirely satisfactory, we can specify the entered phrase, for example by:

“::” - Separating prompts from each other
If you want to separate words, you should put “::” (double colon) after the phrase.

Example:

The “Space ship” prompt will generate images of a spaceship, which is quite logical:

Spaceship

However, the prompt “space:: ship::” will separate these words and MidJourney will perceive them as “Space” and “ship”:

Space and ship

“::x” - prompt gradation
Once we understand how to separate prompts from each other, we can start grading them, i.e. determining which word should be more important in the generated image.