For those of you who haven’t heard about midjourney yet, here is a one line introduction : Midjourney is a text-to-image generation app similar to OpenAI’s DALLE-2 and Stable Diffusion’s DreamStudio , which uses tons of images (around 650+ million) found on the internet — to generate stunning images based on the text prompt provided. It is now in the beta phase — but is adding more features every week .. and the future is yet to be imagined ! ( Currently their AI engine is on v3 )
Anybody who has used the midjourney app is aware that the output will be more colourful and distinctive the more descriptive your writing is. The typical request used by users is “a cowboy wearing a tuxedo on the moon,” which would result in something like this:
To produce predictable and dependable visual outputs, however, there are more choices available than just the simple raw prompt text.
Now getting directly on to advanced options list !
Here are several methods for creating different variations of photos as well as some complex settings for doing so: (details on each of these settings follows the list)
In a nutshell, the keywords of your choice influence your style, and you can choose the types of styles you prefer to pursue. This graphic depicts a general classification of the various style kinds.
Style — keyword
Depending on the type of style you choose, providing a set of supporting prompt keywords connected with “styles” can provide various results. These are some examples of keywords and subtypes based on the genre, artist, or art form you might choose as your style:
using design/genre as style
using artist name as style
You can also specify the artist as your style output . Some samples to the same prompt are shown below
/imagine horse galloping at sunset painting Andy Warhol style
using the rendering/ lighting properties as style
To specify the output resolution, you can use some of the usual keywords like 8K , 4K , photorealistic , ultra photoreal , ultra detailed, intricate details etc.
Or you can go with the standard settings for predictable outputs
hd and quality / --qare two such settings.
/imagine red rose flower --hd
/imagine red rose flower --quality 5
Image as a prompt
When you want consistent results across several photos, supply the URL of an image if you want to get certain outputs that are comparable to that image’s style.
/imagine http://www.imgur.com/Im3424.jpg box full of chocolates
The seed image (located at the URL you provided) and the text questions will both serve as inspiration for the image that is generated.
Several photos can be used as prompts.
You can specify weights to the images ( see the next point )
Weights to Image prompt
If you want your output to be looking more like your prompt image ( see the previous point), give a higher weightage to that image (
keyword : --iw <number>) .
/imagine http://www.imgur.com/ks34f24.jpg chocolates --iw:4
Weights to text prompt
/imagine wild animals tiger::2 zebra::4 lions::1.5
stylize the output
You can add the setting
--s <some number>( denoting style )
The subsequent images, with options for mild and high stylization, are taken from the same prompts.
/imagine firefighters --s 6000
What can be specified. The image output by default is square (1:1 aspect ratio). But, you can alter the aspect ratio if you like a more cinematic look or simply want to create a background for your laptop.
/imagine jasmine in the wild flower --ar 4:3
Use the following example to provide a custom image size
/imagine jasmine in the wild flower --w 600 --h 300
You can choose some of the standard ratios and certain non-standard ones, but not your own bespoke aspect ratios. (These are a few examples)
Filtering out words from your image
--no keyword to discard any unwanted subjects popping in your
/imagine KFC fried chicken --no sauce
chaos — Increase the abstraction
Takes a number from 0 to 100 to increase or decrease the level of abstraction in the subject
/imagine Eiffel tower --chaos 60
Finally, some interesting keywords you can try to use
* Sony Alpha α7, ISO1900, Leica M, = to specify any lens type or camera type * photorealistic , ultra photoreal , ultra detailed, intricate details, = to specify some possible details and a realistic look and less of artistic * unreal = to specify an unreal engine feel * volumetric light , cinematic lighting = to specify some lighting conditions
We discussed how you can have some degree of control over the photographs that your midjourney experiments will produce. Yet the work isn’t nearly finished. The kind people at Midjourney have given a lot of room for your creativity and exploration. Good luck experimenting with various prompts and having fun with your works.