It is possible to have the Midjourney Bot consider two or more separate concepts individually using
:: as a separator. Separating prompts allows you to assign relative importance to parts of a prompt.
Adding a double colon
:: to a prompt indicates to the Midjourney Bot that it should consider each part of the prompt separately.
In the example below, for the prompt
hot dog all words are considered together, and the Midjourney Bot produces images of tasty hotdogs. If the prompt is separated into two parts,
hot:: dog both concepts are considered separately, creating a picture of a dog that is warm.
There is no space between the double colons
Multi-prompts work with Model Versions
Any parameters are still added to the very end of the prompt.
Hot dog is considered as a single thought.
Hot and dog Hot and dog are considered separate thoughts
cup cake illustration
Cup cake illustration is considered together producing illustrated images of cup cakes.
cup:: cake illustration
Cup is considered separately from cake illustration producing images of cakes in cups.
cup:: cake:: illustration
Cup, cake, and illustration are considered separately, producing a cake in a cup with common illustration elements like flowers and butterflies.
When a double colon
:: is used to separate a prompt into different parts, you can add a number immediately after the double colon to assign the relative importance to that part of the prompt.
In the example below, the prompt
hot:: dog produced a dog that is hot. Changing the prompt to
hot::2 dog makes the word hot twice as important as the word dog, producing an image of a dog that is very hot!
3 only accept whole numbers as weights
4 can accept decimal places for weights
Non-specified weights default to 1.
Hot and dog are considered as separate thoughts
Hot is twice as important as Dog
Weights are normalized:
hot:: dog is the same as
hot::100 dog::100, etc.
cup::2 cake is the same as
cup::100 cake::50 etc.
cup:: cake:: illustration is the same as
cup::1 cake::1 illustration::1,
cup::1 cake:: illustration::,
cup::2 cake::2 illustration::2 etc.
Negative Prompt Weights
Negative weights can be added to prompts to remove unwanted elements.
The sum of all weights must be a positive number.
vibrant tulip fields
A range of colored tulips are produced.
vibrant tulip fields:: red::-.5
Tulip fields are less likely to contain the color red.
Weights are normalized so:
tulips:: red::-.5 is the same as
tulips::200 red::-100, etc.
--no parameter is the same as weighing part of a multi prompt to "-.5"
vibrant tulip fields:: red::-.5 is the same as
vibrant tulip fields --no red.