Midjourney is the most popular AI image-generation tool. Here is a beginner’s guide to help you understand Midjourney and be an advanced user.
It is similar to DALL-E and Stable Diffusion by OpenAI and Stability AI, respectively. The company is constantly working on improving its algorithms and releases new model versions every few months. In this guide, we aim to provide you with the latest updates on Midjourney. Currently, one can only access Midjourney via their official Discord server using the Midjourney bot. It can be done either by direct messaging the bot or by inviting the bot to a third-party server. However, we’ll focus on how to use Midjourney for personal use.
Midjourney is a platform that allows you to create cool images and graphics. To access it, all you need to do is follow a few simple steps. Begin by opening an account at Discord, which is completely free. So, let’s start!
Joining Midjourney Discord Server
Here is a guide to joining the Midjourney Discord Channel:
- Login to Discord
- Subscribe to a Midjourney Plan, and choose a subscription plan. Midjourney has a few subscription options, including Basic, Standart, Pro, and Mega plans. (see below for more; check the Midjourney website)
- To start creating images, join the Midjourney Discord server.
- After joining the server, you will see many channels in Midjourney Discord.
- Select any channel labeled #general or #newbie
- Start interacting with the Midjourney Bot on Discord using Commands.
- After checking the Discord server, you can continue creating images in private chat. Right-click on the Midjourney profile picture in the chat and click “Message” to start a private chat with the bot. (see below)
Firstly, I will recommend you just scroll down in a few channels and try to familiarize yourself with MJ. Midjourney is designed to be used by community members so you can see other members’ works on the “general image gen” channel category.
As a paid member, you can send private messages to Discord; you do not need to publish your works on Discord’s main challenge, but using that part is also fun to see what other people are doing there. Also, stay updated by following us and PAACADEMY workshops.
Let’s try to generate an image!
So, firstly, write /image in the direct message box, type a text prompt, and press Enter. You can explain whatever you want to create, but you need to be aware of some points:
Be specific, use descriptive language, and include as many adverbs and adjectives as possible to add nuance and emotion to your prompts (prompt: short text to generate an image). They can describe the color, shape, size, texture, and other characteristics of the object or scene you want to create. Also, Avoid using complex or technical language that may cause Midjourney to become confused.
Here are examples of the prompts:
- Modern house with clean lines, expansive windows, serene interiors, sleek designs, and abundant natural light create an elegant, tranquil ambiance
- Parametric villas with a futuristic, organic design and fluid shapes harmonize with nature
After writing your prompt and sending it to MJ, you just need to wait a few minutes to have your four images.
When you have MJ-generated images, two rows of buttons become available under the images.
Image Selection (U): It helps you separate out your chosen image from the grid.
Image Variation (V): It creates variations of the selected image. Each button generates a new image grid.
Once an image is selected, a more comprehensive range of options will become accessible.
Vary (Strong), Vary (Subtle): These buttons under the image create solid or subtle variations of that image.
Vary (Region): It modifies a selected area within a generated image without changing the entire image.
Upscale: You can use Upscale tools to increase the size of your image.
Zoom Out: It extends the canvas to its original boundaries without changing the content of the original image.
Also, you can interact with the Midjourney Bot on Discord by typing a command. Here are the commands and what they are doing:
/ask: If you have any questions that you want Midjourney Bot to answer, you can simply type “/ask” followed by your question in the message bar. For instance, you may ask a question like “How can I create variations?” The Midjourney Bot will provide you with a step-by-step guide on how to use variations, as shown in the screenshot.
/blend: In Midjourney, you can blend two images using the “/blend” command. Two boxes will appear and you can add your images to blend their styles. Midjourney uses the first image as a base and uses the second image’s color palette and style to generate a new image.
/describe: With the “/describe” command, you can easily upload an image of your choice to Midjourney. After uploading, the platform provides four different descriptions of the image that you can use as prompts to generate a new image in the same style as the original one. This feature can be very helpful in creating a visually consistent set of images for your project.
/fast and /relax: You can switch between /fast and /relax mode by using specific commands. Users subscribed to the Standard, Pro, and Mega plans can create an unlimited number of images each month in Relax Mode. If you use Relax mode sparingly, you will have shorter wait times compared to subscribers who use it more frequently. This priority resets every time you renew your monthly subscription.
/help: Displays important basic information and tips about the Midjourney Bot. You can have information such as how to start, a website, basic commands, extra options, and more.
/imagine: The “/imagine” command is the most important feature of Midjourney. It enables you to generate images on Discord. To use it, simply type “/imagine” in the message box and hit enter. A “prompt” box will appear where you can write a short explanation of the image you want to generate. After sending it to the Midjourney Bot, you will receive your generated image after a short wait.
/shorten: You can use the “/shorten” command to shorten long descriptions. Midjourney Bot offers 5 options.
/settings: View and change the settings for the Midjourney Bot.
/prefer option list: View your existing custom options.
/show: Each image generated by Midjourney is assigned a unique identifier called a Job ID. Job IDs are easily recognizable as they take the form of a string of characters separated by hyphens, such as “9355dcd0-681e-4340-a29c-801e502ae424”. You can find the Job ID in the first part of all image filenames, in the URLs on the website, and as part of an image’s filename.
/subscribe: You can see your subscription model by sending a “/subscribe” message to the Midjourney bot.
/prefer remix: Remix changes the behavior of the variation buttons under image grids. When Remix is enabled, it allows you to edit your prompt during each variation. You just need to click one of the V buttons to see the “Remix Prompt” box.
As mentioned before, a Prompt is a short text phrase to help MJ to generate an image. Unlike basic prompts, More advanced prompts can include one or more image URLs, multiple text phrases, and one or more parameters.
To better understand the structure of Advanced prompts, it is recommended to divide them into three parts: Image Prompts, Text Prompts, and Parameters. Image URLs can be added to a prompt to influence the style and content of the resulting image. Image URLs should always be placed at the beginning of a prompt. The Text Prompt describes what kind of image you want to generate. Lastly, Parameters can be used to change the aspect ratio, models, upscalers, and other settings. Parameters should be placed at the end of the prompt.
Here’s a step-by-step guide on how to upload your image using Midjourney Bot:
- Copy your image and paste it into the chat with Midjourney Bot.
- Press Enter to send your image.
- Depending on your platform, follow these specific steps to obtain the image link
- Discord Desktop App: Right-click on the image and select “Copy Link” (NOT “Copy Message Link”)
- Discord Web App: Click to expand the image, then right-click and choose “Copy image address.”
- Discord Mobile App: Tap and hold on to the image, then select “Copy Media Link.”
- If none of these methods work, don’t worry. You can always click to expand the image, and at the bottom, select “Open in Browser” so you can copy and paste the URL of the image.
Midjourney offers several modes that can be used to apply its default artistic style to your images, you just need to /settings to see default values. The RAW mode is the basic mode and does not apply any style to your images. Stylize low, med, high, and very high set how strongly the default artistic style is applied to your images.
There are three-speed modes available: Turbo mode is the fastest and generates images approximately four times faster, but it comes at twice the cost. Fast mode is the default mode. Relax mode is the free, unlimited mode exclusively available to Pro plan subscribers.
The High Variation Mode and Low Variation Mode determine how much variation is applied when using the V buttons. The Remix mode offers more control but is more complicated. It allows you to change aspects of your prompt when creating new variations.
Also, you can choose between Public mode and Stealth mode to decide whether your generated images are public or private by default.
The Midjourney V5.2 is the latest and most advanced model, which was released in June 2023. This model generates more precise, vivid outputs with improved colors, contrast, and compositions. Additionally, it has a better understanding of prompts than the older versions. You can switch to this model by using the “/settings” command.
Also, Midjourney has partnered with Spellbrush to develop The Niji model. This model is designed to produce anime and illustrative styles with a much deeper understanding of anime, its styles, and aesthetics. It is particularly good at capturing dynamic and action shots, as well as character-focused compositions. To use this model, simply add the parameter –niji 5 at the end of your prompt.
Midjourney is under active development. Stay updated by following us and PAACADEMY workshops.