A "prompt" is a short text phrase that is given to Midjourney Bot to interpret to generate an image. Midjourney Bot breaks down the words and phrases in the prompt into smaller parts, called tokens, which can be compared to the training data and then generate an image. A well-crafted prompt can help you generate unique and stunning images.​

Type composition​

Basic Prompts​

A basic prompt can be just a word, phrase, or emoji.

Advanced Prompts​

More advancedPromptsCan include one or more Image URL,Multiple text phrases, and one or more parameter.

Image Prompts

Image URLs are always placed at the front of the prompt to influence the style and content of the completed result.

See more Image Prompts instructions

Prompt Text Prompt text

A text description of the image you wish to generate. See below for tips and tricks. A well-written tip can help it generate stunning images.

Parameters parameter

Parameters can change the way the image is generated, including adjusting the aspect ratio, model, upscaling, etc. Parameters need to be placed at the end of the prompt.

Read more about Parameters

Prompting Notes​

Prompt Length​

Prompts can be very simple. A single word (or even an emoji!) can generate an image. Prompts that are too short will rely heavily on Midjourney The default style of the tooltip is , so a more descriptive tooltip will create a more unique effect. However, overly lengthy tooltips are not the best. So focus on the main concept you want to convey.

grammar​

Midjourney Bot does not understand grammar, sentence structure, or words the way humans do. In many cases, word choice is also very important. More specific synonyms work better in many cases. For example, big could be replaced with gigantic, enormous, or immense. Use as few words as possible. The fewer words you use, the more impact each word can have. Use commas, brackets, and hyphens to help organize your thoughts, but note that Midjourney Bot cannot reliably interpret these symbols. Midjourney Bot does not consider upper and lower case.

Midjourney Model Version 4 is slightly better than the other models at interpreting traditional sentence structures.

Focus on your needs​

Describe what you want rather than what you don't want. If you ask for a 'no cake' party, your image will most likely contain cake. If you want to make sure a certain object is not in the final image, try using --no parameter Proceed to advanced prompting.

The importance of details

Anything not mentioned may surprise you. Your ideas can be as specific or vague as you want, but anything left out will be randomized. Being vague is a good way to get variety, but you may not get the specific details you want.

Describe any context or details that are relevant to you as clearly as possible. Reference the following:

• Subject: person, animal, character, location, object, etc.
• Medium: photo, painting, illustration, sculpture, doodle, tapestry, etc.
• Environment: indoors, outdoors, on the moon, in Narnia, underwater, the Emerald City, etc.
• Lighting: soft, ambient, overcast, neon, studio lights, etc. (Lighting: soft, ambient, overcast, neon, studio lights, etc.)
• Color: vibrant, muted, bright, monochromatic, colorful, black and white, pastel, etc.
• Mood: Sedate, calm, raucous, energetic, etc. (Mood: Sedate, calm, raucous, energetic, etc.)
• Composition: Portrait, headshot, closeup, birds-eye view, etc. (Composition: Portrait, headshot, closeup, birds-eye view, etc.)

Using collective nouns​

Plural words have a lot of randomness. Try to use specific numbers. "Three cats" is more specific than "cats". Collective nouns are also an option, "a flock of birds" instead of "birds".