Get Better Midjourney Images: A Simple Guide to Advanced Prompts

A Midjourney generated image using Midjourney Automation Suite

Want to create amazing images in Midjourney that look exactly how you picture them? Getting precise results goes beyond just typing a few words. You need to use more advanced prompting techniques.

This guide breaks down five levels of prompting in Midjourney. You'll learn how to move from basic descriptions to powerful prompts that give you detailed control and beautiful results. Stop regenerating images multiple times and start getting closer to your vision on the first try.

Why Go Beyond Basic Midjourney Prompts?

Many people start Midjourney with simple prompts like `/imagine beautiful mountains in the winter`. This works fine for general stock-style images. But if you want something specific – like a sunset in a particular location, precise lighting, or a certain camera look – a basic prompt won't cut it.

Advanced prompting gives you the tools to be much more specific. It allows you to guide the AI to create the exact scene, style, and details you want. This saves you time and gives you much more control over your creative output.

The Five Levels of Midjourney Prompting

There are five main frameworks or levels of prompting you can use in Midjourney:

• Basic Prompt

• Beginner Prompt

• Advanced Prompt

• Multi-Prompt

• Permutation Prompt

Most users might only use the first one or two. Learning the higher levels can greatly improve your results.

1. Basic Prompts

This is the simplest form. It follows the structure: `/imagine [subject] [action] [scene]`. You might add a basic aspect ratio like `--ar 16:9`.

Example: `/imagine beautiful mountains in the winter at the Swiss Alps --ar 16:9`

This is good for getting a general image quickly when you don't need specific details.

2. Beginner Prompts

Beginner prompts add more descriptor words and sometimes basic parameters like `--stylize` or `--chaos`. You might describe the style you want or more details about the subject.

Example: `/imagine giant purple octopus, expansive landscape, top front view, cinematic, Pixar, 32k, white background --stylize 200`

This gives Midjourney more information but still lacks fine-grained control.

3. Advanced Prompts: Getting Detailed Control

This is where you start adding specific instructions about camera type, lighting, shot type, and fine-tuning parameters. An advanced prompt might follow a pattern like:

`[award-winning/professional/cinematic scene] of a [year] [movie genre] [movie name], [shot type], [subject] [scene] [action], shot on a [camera model] [film type], directed by [Name], [emotion] [lighting type], [aspect ratio] --stylize [number] --iw [number] --v [version] --style raw`

Key Elements of Advanced Prompts:

• Prompt Type: Start with a strong descriptor like "award-winning photography" or "professional illustration." You can also specify "product mockup," "graphic design," or "vector."

• Detailed Description: Build on the basic subject, action, and scene. Add specifics like colors, clothing, expressions, or vehicle models (e.g., "Cayenne Red 2023 dually GMC Denali 3500 at4 HD truck").

• Art/Style Directives: Mention specific art styles or photographers if desired.

• Shot Type: This is crucial for composition. Specify how the subject is framed.

• Camera & Film: Naming specific cameras or film types changes the image feel significantly.

• Director/Artist Influence: Asking Midjourney to emulate a director or artist can add their unique style.

• Emotion & Lighting: Tell Midjourney the mood and lighting you want (e.g., "happy," "moody lighting," "dramatic lighting").

• Parameters: Refine the image with parameters:

• `--ar [width:height]`: Set the aspect ratio (e.g., `--ar 16:9`, `--ar 9:16`).

• `--stylize [number]`: Controls how artistic and opinionated Midjourney is (100 to 1000). Higher numbers create more stylized images.

• `--chaos [number]`: Affects the variance in the initial image grid (0 to 100). Higher numbers create more different images.

• `--iw [number]`: Image Weight. Adjusts how much an image or prompt influences the result (0 to 2). Useful when starting from an image.

• `--no [keyword]`: Negative prompt. Tell Midjourney what you explicitly do *not* want in the image (e.g., `--no berries`, `--no fog`).

• `--seed [number]`: Use the seed number from a previous image to generate similar ones.

• `--v [version]`: Specify the Midjourney model version.

• `--style raw`: Uses a different aesthetic, often favoured for photography.

For those looking to generate multiple variations of these advanced prompts efficiently, consider the Midjourney Automation Suite from TitanXT. It can help manage complex prompt structures.

Common Shot Types:

• Extreme Long Shot

• Long Shot

• Medium Shot

• Close-up

• Extreme Close-up

• Full Shot

• Bird's Eye View

• Rule of Thirds Shot

• Silhouette Shot

• Over the Shoulder Shot

• Point of View (POV) Shot

• Low Angle Shot

• High Angle Shot

• Eye Level Shot

• Dutch Angle Shot

• Drone Shot

• Candid Shot

Camera Types to Consider:

Using specific camera types can give your images a particular feel. For example:

• Standard: Canon EOS (portrait, landscape, events), Sony Alpha A7 III (street, sports, wildlife), Hasselblad, Nikon, Panasonic

• Nostalgic: Kodak

• Cinematic: Sony Sonalta, Canon Cinema, Phantom high speed, Blackmagic, Arri Alexa super 16

• Drone: DJI Phantom 4 Pro

4. Multi-Prompts: Weighting Parts of Your Prompt

Multi-prompts use the `::` separator to weigh different parts of your prompt. Midjourney pays more attention to the parts with higher weight (implied weight is 1).

Example: `/imagine fancy chocolate pig statue` (Midjourney focuses on the whole phrase)

Example with Multi-Prompt: `/imagine fancy:: chocolate pig:: statue` (Midjourney gives equal weight to "fancy chocolate pig" and "statue", likely resulting in a statue that is fancy and made of chocolate).

Example with different weights: `/imagine fancy chocolate pig::2 statue` (Midjourney focuses more on the "fancy chocolate pig" part). Experimenting with weighting changes the outcome dramatically.

5. Permutation Prompts: Creating Variations Quickly

This is a powerful time-saver. Use `{}` brackets and commas to include variants directly in your prompt.

Example: `/imagine a cinematic scene of a space movie, {Daytime, Nighttime, lunar eclipse}, shot on a Sony Sonalta camera`

Midjourney will run three separate prompts based on this: one for "Daytime," one for "Nighttime," and one for "lunar eclipse." Each run will still generate the standard four images per prompt.

This technique is great for testing different colors, styles, locations, or subjects without writing separate prompts for each variation. You can include multiple sets of brackets for even more combinations, but remember this uses more credits.

Automating the creation and management of permutation prompts can be handled efficiently by tools like the Midjourney Automation Suite from TitanXT, helping you generate many options with ease.

Bonus Technique: Face Swapping

Midjourney itself doesn't have a built-in face swap feature, but you can use external tools like Insight Face Swap (a Discord bot).

Here’s a basic overview (check the Insight Face Swap website for detailed setup):

• Add the Insight Face Swap bot to your Discord server.

• Upload a photo of the face you want to use by messaging the bot and using the `/saveid` command. Give it a name (e.g., `myface`).

• Find a Midjourney image you want to put the face on.

• Use the `/swapid` command, specify the face ID name (e.g., `myface`), and attach the Midjourney image.

The bot will generate a new version of the image with the swapped face.

Conclusion

Moving beyond basic prompts is key to unlocking Midjourney's full potential and getting the precise images you want. Start by understanding the five prompt levels: Basic, Beginner, Advanced, Multi-Prompts, and Permutation Prompts.

Practice using detailed descriptors, parameters like `--stylize`, `--chaos`, and `--iw`, negative prompts `--no`, specific shot types, camera models, and lighting. Learn to use `::` for weighting and `{}` for generating variations quickly.

With practice, you'll find you can achieve much more control and create truly stunning images tailored to your vision. Keep experimenting and see what amazing art you can create!

To streamline your prompting process and manage complex variations, explore tools like the Midjourney Automation Suite from TitanXT.