002 MidJourney Prompt Guide
002 MidJourney Prompt Guide
Prompting Guide
Test Prompt
When creating text prompts for Midjourney, it's important to remember that anything left unsaid
may surprise you. If you want specific details included in your prompt, you need to be clear
about them. You can be as specific or vague as you want, but if you leave out important context
or details, they will be randomized. Being vague can add variety, but you might not get the
specific details you were hoping for.
When dreaming up your prompt, let your imagination run wild with possibilities! Consider who or
what you want to see in your stunning artwork: a person, animal, character, location, or object?
Next, think about the medium you'd like the artist to use - will it be a photo, painting, illustration,
sculpture, doodle, or tapestry?
Don't forget to set the scene too! Will the artwork take place in a cozy home, the great outdoors,
on the moon, in Narnia, or even underwater? And let's not overlook the importance of lighting - it
can truly make or break the mood of the piece. Do you envision soft, ambient, overcast, neon,
or studio lighting?
Colors can also set the tone, so pick a palette that fits your vision: vibrant, muted, bright,
monochromatic, colorful, black and white, or pastel? Lastly, consider the overall mood you're
trying to convey - is it calm and peaceful, energetic and lively, or something in between? Oh,
and don't forget to think about the composition too - a portrait, headshot, closeup, or birds-eye
view? By being detailed and specific about these elements, you'll be sure to receive an artwork
that perfectly matches your imagination!
In the sections below, I will cover each text prompt attribute and provide examples that you
could use based on your used cases.
- Lighting:
For a photographer, Lightning refers to the ability to use lights effectively, to illuminate and bring
out the features in an image or specifically the object in the image.
-Artists:
An artist's name can be used as a prompt for an AI image generator by providing the name of a
specific artist as input to the algorithm. The AI will then use the style and techniques of that
particular artist to generate an image. This is can be useful for you when you want to
experiment with different styles and techniques without having to learn them all themselves.
Color:
In the context of using color options as a prompt for Mid Journey, "color" refers to the specific
hues or shades that can be used to generate an image. Color plays a significant role in the
visual arts, and it is often used to evoke specific emotions and convey meaning in an image. By
using color options as a prompt for an AI art generator, artists can experiment with different
color schemes and explore how different colors can affect the overall mood and meaning of an
image.
Additionally, the choice of colors can also affect the perceived mood of an image. Bright and
bold colors can create a feeling of energy and excitement, while subdued colors can create
a
feeling of calm and tranquility. By using color options as a prompt, an artist can experiment with
different moods and emotions in their images.
Theme/Styles:
In Context of Photo generation style/theme refers to the ability to apply a specific aesthetic or
visual theme to the generated image. A few examples below:
Camera:
In the context of photo generation, the camera options can be used to simulate the effects of
different types of cameras and lenses in the final output image.
● Panorama: A style that captures a 360-degree image of the scene, with a wide and
expansive view that can create a sense of immersion and grandeur.
● DSLR: A style that replicates the realistic look of an image taken with a DSLR camera,
with sharp focus, a shallow depth of field, and a balanced exposure.
● Magnification: A style that simulates the look of a magnifying lens, with a focus on the
texture and detail of the image. This can create a macroscopic effect, emphasizing the
small details of the scene.
● Microscopy: A style that simulates the look of an image captured under a microscope,
with a focus on the microscopic details of the image. This can create an abstract and
surreal effect, revealing patterns and structures that are invisible to the naked eye.
● Pinhole Lens: A style that simulates the look of an image captured through a pinhole lens,
with a soft and dreamy quality to the image. This can create a vintage and nostalgic
effect, with a focus on mood and atmosphere.
● Bird's Eye View: A style that captures the scene from a high-angle perspective, looking
down on the subject from above. This can create a sense of distance and detachment,
or a sense of grandeur and awe.
● Worm's Eye View: A style that captures the scene from a low-angle perspective, looking
up at the subject from below. This can create a sense of drama and tension, or a sense
of vulnerability and intimacy.
● Tilt-Shift: A style that simulates the look of a miniature model or diorama, with a selective
focus that creates a sense of depth and dimension.
● HDR: A style that combines multiple exposures of the same scene to create an image
with a high dynamic range, with details in both the highlights and shadows. This can
create a realistic and immersive effect, with a sense of depth and detail.
● Fish-Eye: A style that uses a wide-angle lens to create a distorted and exaggerated
perspective, with a curved and distorted view of the subject. This can create a
surreal and otherworldly effect, with a focus on mood and atmosphere.
● LIDAR: A style that uses LIDAR technology to capture a 3D map of the scene, with a
focus on depth and geometry. This can create an abstract and futuristic effect, with a
sense of dimension and spatial awareness.
● Thermal: A style that uses thermal imaging to capture the heat signature of the scene,
with a focus on temperature and contrast. This can create a stark and dramatic effect,
with a focus on the interplay of light and shadow.
● Hyperspectral: A style that uses hyperspectral imaging to capture a range of frequencies
beyond the visible spectrum, with a focus on color and texture. This can create a surreal
and otherworldly effect, with a focus on the interplay of light and material.
● Multispectral: A style that uses multispectral imaging to capture a range of wavelengths
within the visible spectrum, with a focus on color and contrast. This can create a vibrant
and colorful effect, with a focus on the interplay of light and color.
● 360 VR: A style that captures a 360-degree virtual reality image of the scene, with a
focus on immersion and interactivity. This can create a sense of presence and spatial
awareness, with a focus on interactivity and exploration.
● Augmented Reality: A style that incorporates digital elements into the image, using AR
technology to add virtual objects and effects to the scene. This can create a playful and
interactive effect, with a focus on creativity and imagination.
● Stereoscopic: A style that uses stereoscopic imaging to create a 3D effect, with a focus on
depth and dimension. This can create a realistic and immersive effect, with a sense of
spatial awareness and presence.
● Motion Capture: A style that uses motion capture technology to capture the movement of
the scene, with a focus on motion and dynamics. This can create a dynamic and kinetic
effect, with a sense of energy and motion.
● Time-Lapse: A style that captures the scene over an extended period of time, using
time-lapse technology to condense hours or days into a single image. This can create a
surreal and dreamlike effect, with a sense of time and motion.
● High-Speed: A style that captures the scene at a high frame rate, using high-speed
technology to slow down and capture fast-moving action. This can create a dramatic and
intense effect, with a focus on the interplay of motion and time
Camera Angles:
Camera angles refer to the position of the camera in relation to the subject being photographed
● Eye Level: A camera angle that captures the scene from the perspective of a person's
eye level. This can create a sense of familiarity and relatability, with a focus on the
subject's point of view.
● High Angle: A camera angle that captures the scene from above, with a focus on the
subject's position in the environment. This can create a sense of vulnerability or
powerlessness, with a focus on the subject's relationship to the surrounding space.
● Low Angle: A camera angle that captures the scene from below, with a focus on the
subject's height and position in the environment. This can create a sense of power or
authority, with a focus on the subject's dominance over the surrounding space.
● Rear View: A camera angle that captures the scene from behind the subject, with a focus
on the subject's posture and body language. This can create a sense of mystery or
detachment, with a focus on the subject's hidden or internal thoughts.
● Front View: A camera angle that captures the scene from the front of the subject, with a
focus on the subject's facial expressions and emotions. This can create a sense of
intimacy or confrontation, with a focus on the subject's external reactions.
● Closeup: A camera angle that captures the subject's face or body in close proximity, with
a focus on detail and texture. This can create a sense of intensity or intimacy, with a
focus on the subject's physical presence.
● Full Body: A camera angle that captures the subject's entire body, with a focus on the
subject's posture and movement. This can create a sense of dynamism or stillness, with
a focus on the subject's physicality and motion.
● Oblique: A camera angle that captures the scene at a diagonal or slanted angle, with a
focus on the subject's relationship to the environment. This can create a sense of tension
or instability, with a focus on the subject's position and movement in space.
● Dutch Angle: A camera angle that captures the scene at a tilted or canted angle, with a
focus on disorientation or imbalance. This can create a sense of unease or chaos, with a
focus on the subject's psychological state.
● Aerial View: A camera angle that captures the scene from above, with a focus on the
subject's position in the larger environment. This can create a sense of scale and
perspective, with a focus on the subject's relationship to the wider world.
Camera Types:
Images could be further enhanced by specifying the camera type that you want the MidJounrey
to use as a reference to generate images.
Material:
Specific materials can also influence the composition of the image, some of the options worth
trying are below:
Aluminum, Brick, Bronze, Carbon Fiber, Cardboard, Cellulose, Ceramic, Cotton, Fabric, Fiber
Optic, Foil, Gasoline, Glass, Gold, Gummies, Latex, Leather, Magma, Metallic, Nickel, Nylon,
Paper, Plastic, Quartz, Sharink Wrap, Skin, Slime, Wooden, Yarn
Artistic Mediums:
In the context of images, the term "medium" refers to the material or format used to create,
store, or display an image. This can include traditional art media such as canvas, paper, or
paint, as well as digital media such as computer screens, printers, or digital files. The medium
used can have a significant impact on the look and feel of the image, as different materials and
formats can create different textures, colors, and effects.
● Mosaic: Art made by assembling small pieces of colored glass, stone, or tile. ●
Collage: Art made by gluing together pieces of paper, fabric, or other materials. ●
Tapestry: Woven textile art made on a loom with designs or images.
● Etching: Intaglio printmaking technique using acid to create lines in a metal plate.
● Charcoal drawing: Art made with charcoal, often used for sketches and portraits. ●
Watercolor: Painting with water-based pigments on paper.
● Wood carving: Sculpture made by carving wood with tools.
● Glassblowing: Art of forming molten glass by blowing air into a tube.
● Calligraphy: Decorative handwriting with a pen or brush.
● Chalk pastel drawing: Art made with soft pastel chalks on paper.
● Digital art: Art made using digital technology such as computer programs.
● Graffiti: Art made by writing or drawing on public surfaces.
● Performance art: Live art performance, often incorporating theater or dance.
● Metalwork: Art made by forging, casting, or welding metal.
● Embroidery: Decorative stitching on fabric with thread or yarn.
● Pottery: Art of making objects from clay, fired at high temperatures.
● Engraving: Printmaking technique using incised lines on a metal plate.
● Printmaking: Art of creating multiple prints from a single design.
● Origami: Art of folding paper into decorative shapes or figures.
● Sand art: Art made by arranging sand into patterns or designs.
● 3D printing: Art made using a 3D printer, which prints three-dimensional objects.
● Paper mache: Art made by shaping and sticking layers of paper and adhesive. ●
Beadwork: Art made by stringing beads together into patterns or designs. ●
Leatherwork: Art of crafting objects from leather, often used for belts or bags. ●
Mixed media art: Art made by combining different materials or techniques.
Emotions:
Defining emotions in the context of an image refers to the portrayal of emotions through the use
of visual elements such as color, composition, and facial expressions. The emotions conveyed
in an image can evoke a particular mood or feeling in the viewer, and can be an important
aspect of the image's overall impact.
Joyful, serene, playful, mysterious, melancholy, powerful, nostalgic, intense, romantic, hopeful,
Energetic, tranquil, contemplative, dreamy, dramatic, whimsical, pensive, optimistic,
adventurous, enigmatic.
—----------------------------------------------------------------------------------------------------------------------------
Parameters
Aspect Ratio
Aspect ratio is a term used to describe the proportional relationship between the width and
height of an image, video, or display screen. It is expressed as a ratio of the width to the height,
typically separated by a colon.
For example, a 4:3 aspect ratio means that the width of the image is four units and the height is
three units. This is a common aspect ratio used in older television screens. A 16:9 aspect ratio
is a wider, rectangular shape and is commonly used in modern widescreen displays.
Aspect ratio is an important consideration when creating or viewing content, as it can affect the
overall composition and visual impact of an image or video. Different aspect ratios can also
affect how content is displayed on different devices, and may require adjustments to ensure that
the content is properly formatted for different screens.
Here are some popular aspect ratios and their common use cases:
1:1 - Square aspect ratio, often used for social media platforms like Instagram and
Facebook. 4:3 - Standard aspect ratio for older TV screens and digital cameras.
16:9 - Widescreen aspect ratio used for modern TVs, monitors, and laptops, as well as video
content on YouTube, Netflix, and other streaming platforms.
21:9 - Ultra-wide aspect ratio used for some computer monitors and cinema screens to provide
an immersive viewing experience.
2.39:1 - CinemaScope aspect ratio used for widescreen movies, which provides a cinematic
experience with a wide view.
5:4 - Common aspect ratio for computer monitors, especially in the past. 3:2
- Used by many digital cameras to capture images in a 35mm film format.
Different aspect ratios are suited for different purposes and media formats, and it is important to
choose the right one for the content you are creating or viewing.
Aspect ratio and resolution are two important but distinct concepts in the field of digital imaging.
Aspect ratio describes the proportional relationship between the width and height of an image or
display screen, and is typically expressed as a ratio such as 4:3 or 16:9. Resolution, on the
other hand, refers to the number of pixels that make up an image or screen, and is typically
expressed as the total number of pixels horizontally and vertically, such as 1920x1080 or
3840x2160. Although these concepts are related, they describe different aspects of digital
content. Aspect ratio determines the overall shape and composition of an image or screen,
while resolution determines the level of detail and clarity of the visual content. Understanding
these concepts is important for properly creating, displaying, and manipulating digital images
and video content.
Quality
The parameter "--quality" or "--q" is responsible for determining the time spent generating an
image. Higher quality settings produce more details, but also take longer to process and utilize
more GPU minutes per job. It is important to note that the quality setting does not affect the
resolution of the image. Therefore, while higher quality settings may lead to a more detailed
image, they also come at the cost of higher processing time and resource usage.
Stop
You can use the "--stop" parameter to prematurely terminate a Job before it completes its
processing. It is important to note that stopping a Job before it reaches a higher percentage of
completion may result in blurry and less detailed output. If additional processing is required to
achieve a higher quality output, it may be necessary to restart the Job and complete the full
processing.
Stylize
The Midjourney Bot has been trained to generate images that prioritize artistic color,
composition, and forms. The level of artistic influence on the output can be controlled by
adjusting the "--stylize" or "--s" parameter. Lower stylization values will result in images that
more closely resemble the original prompt but with less artistic flair. Conversely, higher
stylization values will result in more artistic images but with a weaker connection to the
original
prompt. It is therefore important to carefully balance the stylization value according to the
desired output and the specific use case.
Seed
The Midjourney bot is capable of generating stunning images by utilizing a unique method of
producing visual noise. This process begins by using a seed number to generate a field of
noise, which is then used to initiate the image creation process. These seed numbers are
generated randomly for each image, but they can also be explicitly specified using the "--seed"
or "--sameseed" parameter (which is applicable on V3 and below).
The seed number feature in Midjourney is particularly useful when generating multiple images
with similar characteristics, as it allows the bot to produce images that share visual similarities.
By using the same seed number and prompt, similar final images can be created. This feature is
especially helpful when a series of images with a consistent look and feel is desired.
In version 4 of the Midjourney model, using the same seed number will produce the exact same
image again. This means that any images generated using the same seed and prompt will be
identical. This feature can be advantageous when it comes to replicating the same image
multiple times, or when a specific image needs to be recreated with precision.
On the other hand, Model Versions 1, 2, 3, test, and testp will produce similar, but not identical,
images when the same seed is applied to the image. This is because these models use a
slightly different method for generating visual noise compared to version 4. While the images
produced using the same seed and prompt may not be identical, they will still share similar
visual characteristics, which can be useful for generating a series of related images.
Overall, the seed number feature in Midjourney provides a powerful tool for generating images
with consistent visual characteristics. Whether you want to create a series of related images,
replicate the same image multiple times, or generate a specific image with precision, the ability
to specify the seed number can be invaluable.
Chaos
The --chaos or --c parameter controls the level of variation in initial image grids produced by the
Midjourney AI. The higher the --chaos value, the more unusual and unexpected the results and
compositions will be. The default --chaos value is 0, and --chaos accepts values between 0 and
100. When using low --chaos values, the initial image grids are slightly varied each time a job is
run, resulting in reliable and repeatable outcomes. However, using extremely high --chaos
values will produce initial image grids that are varied and have unexpected compositions or
artistic mediums each time a job is run. As an example, using the prompt "cat jug hybrid"
and
setting the --chaos value to 100 will produce highly unpredictable and imaginative outcomes.
This feature provides a level of excitement and creativity for users who want to experiment with
new and unexpected concepts.
Tile
By utilizing the "--tile" parameter, the Midjourney bot can produce images that are suitable for
creating seamless patterns for fabrics, wallpapers, and textures. The resulting images can be
tiled repeatedly, without any noticeable seams or borders, to produce visually cohesive and
continuous patterns. This feature can be particularly useful in various design applications, where
seamless patterns are frequently required.
Advanced Prompts
Text Weights
In the Midjourney Bot, the double colon "::" can be used to divide a prompt into separate parts,
and a number can be added after it to indicate the relative importance of each part of the
prompt. For instance, if the prompt is "butterfly," the bot will generate an image that relates to
this concept. However, if the prompt is modified to "butter::2 fly," the importance of the first part
of the prompt, "butter," is doubled relative to the second part, "fly." As a result, the bot generates
an image that emphasizes the butter part of the prompt, perhaps creating an image of a butterfly
made of butter or set against a buttery background. This feature enables users to specify the
relative importance of different parts of a prompt, resulting in more nuanced and tailored images
that align with their creative vision.
Image Prompting
Using images as part of a prompt in the Midjourney Bot can significantly influence a Job's
composition, style, and colors. Image prompts can be used either alone or in combination with
text prompts, and users can experiment with different image styles to achieve the most exciting
results. Incorporating images in prompts provides users with more creative freedom to express
their vision and desired style. By combining images with text prompts, users can create more
specific and nuanced prompts that enable the bot to produce more accurate results. The
Midjourney Bot can recognize and interpret the content of the image prompts, such as the
colors, shapes, and forms present in the images, and apply them to generate visually stunning
and unique images.
To add images to your prompt, simply type or paste the web address of the image you want to
use. Make sure the image is stored online and has an extension such as .png, .gif, or .jpg. Once
you have added the image addresses, you can add any additional text and parameters to
complete your prompt. The Midjourney Bot will interpret the content of your image prompts,
including colors, shapes, and forms, and incorporate them into the generated image. By
experimenting with different images and prompts, you can express your creativity and generate
visually stunning images.
Niji Model of MJ
The Niji model, named after the Japanese word for both "rainbow" and "2D," is a new image
generation model specifically designed for anime and illustrative styles. The Niji model has been
fine-tuned with a careful eye to produce high-quality anime-style images, with a vast amount of
knowledge about anime, anime styles, and anime aesthetics. It excels at producing dynamic
and action shots, as well as character-focused compositions.
To get started using the Niji model, there is an anime style formula that can be added to
prompts. This formula is the most general modifier available, called the "anime style" modifier,
and is denoted by the flag --niji. It is a great starting point for creating anime-style images with
the Niji model, and from there, users can experiment with different anime styles to create unique
and interesting images.
One anime style that can be incorporated into prompts is the Chibi style, also known as the
super-deformed style. This style features characters with exaggerated, childlike proportions,
with large heads and small bodies. It is often used to add humor or levity to a scene and has
become a popular style in anime and manga. Some examples of anime that utilize the Chibi
style include "Pani Poni Dash!" (2005), "Lucky Star" (2007), and "Hetalia: Axis Powers" (2009).
Another type of anime that can be incorporated into prompts is Gakuen, which refers to anime
where the story takes place in a school setting. Gakuen anime often focuses on the
relationships and conflicts of its school-aged characters and may include romance, drama, or
comedy elements. Examples of Gakuen anime include "Love, Chunibyo & Other Delusions"
(2012), "The Pet Girl of Sakurasou" (2012), and "Boarding School Juliet" (2018).
The Niji model's focus on anime and illustrative styles is a significant development in image
generation, as it allows artists and designers to create high-quality anime-style images quickly
and easily. With the vast amount of knowledge about anime, anime styles, and anime
aesthetics, the Niji model has the potential to revolutionize the anime and manga industry by
allowing creators to quickly and easily generate high-quality images for their projects.
In addition to the anime style modifier, the Niji model also includes a variety of other parameters
that can be used to customize and fine-tune the generated images. These parameters include
the aspect ratio of the image, the image size, and the number of images to generate. Users can
also specify the seed to be used in the random number generator, allowing them to recreate
images that they have generated previously.
Overall, the Niji model is an exciting development in the world of image generation, particularly
for those interested in creating anime-style images. Its fine-tuning for anime and illustrative
styles, along with the vast amount of knowledge it possesses about anime, anime styles, and
anime aesthetics, make it a powerful tool for artists and designers looking to create high-quality
anime-style images quickly and easily. With the ability to experiment with different anime styles
and customize the generated images, the Niji model has the potential to revolutionize the anime
and manga industry and change the way artists and designers approach creating anime-style
images.
HAPPY PROMPTING ��