The Ultimate Beginner's Guide to Midjourney in 2024

Midjourney has evolved significantly, especially with Version 6. This guide will break down the changes, explain the features, and preview what's coming in the next year.

Prompting in Version 6

Prompting has changed. Use `/imagine` to start. Version 5 prompts like "a blue sports car in Monaco, pink road, overcast weather" often missed details. Version 6 requires more detail.

Example of Prompting Differences

Instead of the old way, try: "wideangle product photo of a sports car in Monaco, the sports car is bright blue, the road is painted pink, the weather is overcast." The AI now picks up more details. Midjourney fills in any details you don't mention. If you're moving from version 5, rewrite your prompts to suit the new bot.

Adding Detail Incrementally

Start with a simple prompt and add details one at a time to see how the AI responds. For example, begin with "a full body shot of a man wearing a Celestial Crown." Then, add "blue sunglasses," "red earrings," and so on. This helps you test the prompt's coherence.

Prompting Specificity

Be specific with details. Instead of general terms, specify clothing types, patterns, or materials. There are nearly endless possibilities now, and the AI will only become more powerful. If the bot isn't getting what you want, try a different way to write the prompt. There's no exact formula.

Adding Text to Images

You can add text to images by putting words in quotation marks. For example: "an old timey computer screen that says hello world." Text generation can be hit or miss, and is easier with short words. Keep in mind that an update is expected soon that will improve text generation.

Style Raw Parameter

Try using `--style raw` for a more literal interpretation of your prompt. This is useful for longer prompts or when the bot isn't listening. It may also improve text generation. For example, "a neon sign on a dark black wall background --style raw".

Specifying Locations

Specify where you want elements to appear in the picture, like "a beautiful woman with Midjourney tattooed across her forehead, V6 tattooed on her left cheek." Note that the bot interprets "left" from its perspective, not yours.

Multi-Prompting

When the bot isn't getting what you want, use multi-prompting. Split your prompt using two colons and add weighted values. For example: "there is a double exposure effect with glitchy elements on the photo:: glitch core Double Exposure 0.5". This adds a little extra emphasis to certain elements.

Want to take your Midjourney creations to the next level? Check out the Midjourney Automation Suite from TitanXT for tools to enhance your workflow.

Upscaling Images

When you find an image you like, you have several options. Hitting the U button underneath the generation upscales the image, giving you four options underneath the picture:

• Very Subtle: Creates four more images based on the picture, with very subtle tweaks. Good for creating slideshows with small changes.

• Very Strong: Creates images that are quite different, but keep the general ingredients the same. Useful for fixing small imperfections.

Remix Mode

In your settings, turn on remix mode. Every time you hit a variation button, a new prompt box appears, allowing you to regenerate the image with a changed prompt. Change details like time of day or eye color. This can result in some of the coolest images you'll ever see.

Upscale Subtle and Creative

Beneath an upscaled image, you have two more upscale options:

• Upscale Subtle: Creates a higher resolution version of your picture.

• Upscale Creative: Changes your image, reinterpreting your prompt and making corrections.

Parameters

Parameters are things you can add to your prompt that will change the outcome. They move you away from the default look of Midjourney.

Stylize Value

The stylize value determines how beautiful your image will be. It balances how much the bot listens to your prompt versus its creative freedom. The values range from 0 to 1,000, with a default of 100. Use `--stylize` or `--s` to trigger it.

• Lower Values (below 100): Midjourney will follow your prompt more closely. Results in ugly images.

• Higher Values (above 100): Results in more beautiful images, but might come at the cost of prompt accuracy.

Chaos Value

The chaos parameter brings variety into your grid. Trigger it by adding `--chaos` or `--c` to your prompt, followed by a value between 0 and 100 (default is 0). A little chaos can get you cool pictures, but a lot of chaos means you have no idea what to expect.

Weird Parameter

The weird parameter makes your images look more weird. Trigger this by using `--weird` or `--w` inside of your prompt, with a numeric value between 0 and 3,000 (default is zero). It sticks to your prompt more closely than the stylize value or chaos, but the results may not be beautiful.

Saving Parameter Shortcuts

To save a recipe of parameters you like, use `/prefer option set` in the prompt box on Discord. Create a name for your shortcut, then click the plus button to input the values. To trigger the shortcut, use `--yourshortcutname` in your prompt.

Prefer Suffix

To automatically add parameters to the end of all your prompts, use `/prefer suffix`. This is best used for aspect ratios, which can be easily overridden in individual prompts.

Automate your Midjourney experience with TitanXT! Streamline your creative process and bring your visions to life faster. Visit TitanXT today!

Image Prompting

To use image prompting on Discord, click the plus button beside the prompt box and upload a file. Then, drag the picture into the prompt box. You can combine multiple pictures and add words. Alternatively, use `/blend` to blend up to five images without adding words.

Describe Feature

Use `/describe` to upload an image, and Midjourney will give you four prompt descriptions based on that picture. This is an older version of describe, and an updated version is coming soon. To recreate a picture, run it through the describe function, then copy the image address and paste it inside the described prompt.

Features Coming Soon

There are many more features expected to come to Midjourney soon:

• Inpainting: Change individual parts of an image.

• Zoom Out: Zoom out at a 1.5x or 2x scale, or use custom zoom to change the aspect ratio.

• Panning: Expand the frame to the left, right, up, or down, potentially with remix mode turned on to change the prompt.

• Style Tuning: Create a tuning test to choose which style best suits what you were looking for.

Niji Journey

Niji Journey is a subsidiary of Midjourney focused on anime. Expect version six of Niji Journey sometime soon. It includes various styles like expressive, scenic, and cute.

Midjourney's New Website

Midjourney has a redesigned website with impressive features:

• Explore Page: Browse popular and recent creations from the community.

• My Images: View your entire gallery and download images. The search is instantaneous.

• Rate Images: Help train the bot by rating which images you like more.

• Creation: Enabled for those who have made 10,000+ images, offering new setting sliders.

Website Updates in the Future

Expect these website updates:

• Creation enabled for everyone.

• Social features similar to Discord.

• Organizational features like folders, collections, and personal workspaces.

Overall Tools to Expect

Here are overall tools to expect in the future for Midjourney:

• Consistency tools for both style and character.

• Storytelling tools.

• Video generation.

• 3D generation.

Version 7 and Version 8 should be out before the end of 2024. There will be constant updates with a date-based release cycle.

Conclusion

This guide covered everything you need to know about Midjourney in 2024. As you continue to explore Midjourney's capabilities, consider how the Midjourney Automation Suite from TitanXT can further streamline your creative process. Visit TitanXT to learn more!