Midjourney Explained: ALL Midjourney Features Breakdown
- AI Catalyst
- June 24, 2024
- Gallery
- 0 Comments
Prompt
A prompt is a text phrase describing an image you want to create. You need to type your prompt using the /Imagine command into the Midjourney bot and Midjourney will generate an image. A prompt can be any word, sentence, single letter, or even an emoji.
/describe
The /describe command allows you to upload an image and gives you 4 textual descriptions of that image which can be used as prompts to generate new images similar to the original one.
--no
The –no parameter tells the Midjourney Bot what not to include in your image. For example, if you write “no cake” in your prompt, the image will most likely include a cake but if you use the no parameter a cake will not appear in your image.
Image Prompt
You can use images as part of a prompt to influence a Job’s composition, style, and colors. Image prompts can be used alone or with text prompts.
Multi Prompt
Multi Prompting allows you to blend multiple concepts using :: as a separator. It assigns relative importance to the concept in the prompt, helping you control how they are blended together.
Remix
Remix Mode allows changing prompts, parameters, model versions, or aspect ratios between variations. It’s activated using the /prefer remix command.
/shorten
The /shorten command analyzes your prompt, suggests edits highlighting essential terms, and removes unnecessary words.
/blend
The /blend command allows you to upload 2–5 images and merge them into a new image using their aesthetics and concepts.
/info
The /info command shows you basic information such as your current queued and running jobs, subscription type, renewal date, Remaining fast GPU hours, and more.
/settings
The /settings command shows buttons for common options like Model Version, Stylize value, Variation value, and more.
/show
You can use the /show command with the unique Job ID to move a job to another server or channel, revive a lost job, refresh an old job, and more.
/subscribe
/Subscribe opens a webpage where you can manage your Midjourney subscription.
--ar
The –aspect or –ar parameter changes the aspect ratio of the generated image. An aspect ratio is the width-to-height ratio of an image. It is typically expressed as two numbers separated by a colon, such as 16:9, 9:16, or 4:3
--chaos
The –chaos or –c parameter produces more unusual and unexpected results in compositions in images. Higher and lower values are available.
--q
Quality: The –quality or –q parameter changes how much time is spent generating an image. Higher-quality settings take longer to process and produce more details but they consume more of your GPU hours. The default value is 1 and it only accepts values: .25, .5 and 1.
--r
Repeat: The –repeat or –r parameter runs the same prompt multiple times. There are different limits for each subscription plan.
--w
Weird: –weird or –w parameter adds weird and unusual details to your generated images, resulting in unique and unexpected outcomes. It accepts values from 0 to 3000.
Seed Numbers
Seeds: Seed numbers are generated randomly for each image. If you use the same seed number and prompt, you will get similar final images.
--stop
–stop parameter finishes a Job partway through the process. It accepts values from 10 to 100 with 100 being a default value and 10 being the blurriest image possible.
--style
The –style parameter replaces the default aesthetic of some Midjourney Model Versions.
--s
Stylize: The –stylize or –s parameter influences how strongly Midjourney applies its training data to your image. Low values produce images that closely match the prompt but are less artistic and high values produce more artistic images but they are less connected to your prompt.
/tune
Style Tuner: /tune command generates various sample images showing different visual styles based on your prompt. You can choose images you like, which will then customize the appearance of your next images.
--sref
Style Reference: Style reference (–sref) lets you use an image as a reference for your prompt. Compared to the image prompt, the style reference focuses more on transferring the overall aesthetics and not the objects.
--tile
Using the –tile parameter you can generate images that can be used as repeating tiles to create seamless patterns.
--v
Version: You can switch Midjourney model versions using the –version or –v parameter. The latest version is the default value.
Vary Region
The Vary Region button opens an editor where you can select and regenerate specific parts of an upscaled image.
Pan
The Pan button allows you to expand the image in a chosen direction without changing the content of the original image.
Zoom
The Zoom button allows you to zoom out the upscaled image without changing the content of the original image. You can also change the aspect ratio of the image and complement your initial prompt.
Upscale
Upscale buttons upscale the selected image from its grid. You can then use the Creative or Subtle Upscale tools to increase the size of your image. The subtle Upscaler doubles the size of your image and keeps details very similar to the original whereas the Creative upscasler also doubles the size of your image but adds new details to the image.
Variations
The V buttons under the image grid create subtle or strong variations of the selected image. You can also create variations using the buttons that appear under upscaled images. Subtle variations produce new images that retain the main composition of the original image but introduce subtle changes to its details. Strong variations will produce new images that may change the composition.
Permutation
Permutation Prompts allow you to quickly generate variations of a prompt. You need to include lists of options separated with commas, within curly braces {} in your prompt, and it will create multiple versions of a prompt with different combinations of those options.
Remaster
With The Remaster feature you can upscale images generated using earlier Midjourney Model Versions. It will create a new grid of images using the latest Midjourney Model.
--video
The –video parameter creates a short movie of your image grid being generated.
/stealth
Pro and Mega plans subscribers have access to Stealth Mode, which prevents images from being visible to others on the Midjourney website.
/turbo
Turbo mode: You can switch to Turbo Mode if you want your images to be generated 4 times faster, but they will consume twice as many subscription GPU minutes as a typical Fast Mode Job. Relax Mode does not cost any GPU time, but Jobs will be placed into a queue based on how much you’ve used the system.
Niji model (--niji)
The Niji model is a special Midjourney model specifically tuned for Eastern and anime aesthetics. It is best suited for creating dynamic and action shots and character-focused compositions. You need to add the –niji 6 parameter to your prompt or select the Niji model from the settings.
Website
If you have an active Midjourney subscription plan you can access the alpha version of the website to browse, download, and rate images created by the community, copy prompts, see your images, and generate images outside of Discord.
--cref
Consistent Character: Similar to the style reference feature, the newly introduced –cref parameter allows you to use the character in the reference image in your Midjourney images. You need to type cref after your prompt and add a link to an image of a character. You can use cw parameter to modify reference strength from 0 to 100.
--sv
Versioning: The versioning feature allows you to choose the version of your style reference algorithm using the –sv parameter. There are 4 available options with 4 being the default algorithm.
--p
Personalization: Personalization allows you to create your own unique style by ranking pairs of images on the Midjourney website and using the –p parameter for your next prompts. It changes the appearance of your next images, taking your choices into account when choosing between two images, and works similarly to the Style Tuner.
