Midjourney's Latest V7 Update: What's New and How to Use It

Midjourney recently rolled out significant updates to its V7 model. These changes bring better image quality and powerful new tools. If you're using Midjourney, understanding these updates will help you create better images. Let's look at the key new features and how they work.

Getting Started with V7

First, make sure you are using the V7 profile. If you are new to V7 or haven't set it up, you might need to personalize your profile. This involves selecting 200 images you like in the personalization section to train your profile on your preferred style.

Improved Image Quality in V7

The V7 update brings general improvements to all generated images. Expect better accuracy, more natural-looking hands, and more realistic body proportions. Simply using the `--v 7` parameter will give you noticeably better results compared to previous versions of V7.

Explore the New `--xp` Parameter

A new parameter called `--xp` or `--experience` is now available. This parameter makes your images more detailed and dynamic. You can combine it with `--stylize` to influence the tone.

• The `--xp` value ranges from 0 to 100. The default is 0.

• Add `--xp` at the very end of your text prompt.

• Testing shows that lower values like `--xp 5` might stay truer to your original prompt in terms of color and pose compared to higher values like 25 or 50. Higher values can sometimes lean more towards the standard Midjourney aesthetic rather than your specific request.

• Experiment with different `--xp` values to see what you like best for different prompts.

Finding the Right `--xp` Balance

Experimentation is key with `--xp`. Try values like 5, 25, and 50 to see how they affect the level of detail and how closely the result matches your prompt. What works well depends on the subject and style you are aiming for.

Managing many experiments with new parameters like `--xp` and `--ow` can become time-consuming. Tools designed for Midjourney automation can help you organize and run tests more efficiently. Check out the Midjourney Automation Suite from TitanXT to streamline your workflow.

Character Consistency with Omni Reference (`--or`)

One of the most anticipated features is Omni Reference, using the `--or` parameter. This lets you use an image reference to create characters that stay consistent across different images.

To use it, simply drag your reference image into the prompt area.

Controlling Force with Omni Weight (`--ow`)

The `--ow` (Omni Weight) parameter controls how strongly Midjourney follows your reference image. The default value is 100, but you can set it from 0 to 1000.

• A higher `--ow` number means Midjourney will try harder to match the reference image, including face and clothing.

• A lower `--ow` number gives Midjourney more freedom. This is useful if you want to use the character's face but change their style, like turning a photo into anime (try `--ow 25`).

• For a very close match to both face and clothes, use higher values like `--ow 400` to 1000.

Experimenting with Omni Reference Results

Using Omni Reference requires some trial and error. Results can vary depending on the prompt, the reference image, and the `--ow` value. Sometimes the face matches but the clothes don't, or vice versa. Combining `--or` and `--ow` with other parameters like `--raw` (for realistic style) and `--stylize` adds more variables to test.

The developers are still working on perfecting Omni Reference, so expect it to improve over time. Testing different combinations of `--ow` and `--stylize` values is necessary to get the desired consistent character.

If you find the process of running multiple tests for character consistency or parameter variations tedious, consider using an automation tool. The Midjourney Automation Suite by TitanXT can help you manage these experiments and generate variations automatically.

Training Your Midjourney Profile

Did you know you can train Midjourney to understand your style preferences? You can personalize your profile by liking images in the explore section. When you like images there, it helps Midjourney learn your taste and influence your future generations. Liking your own generated images, however, does not train your profile in this way.

Future Updates: The Image Editor

Midjourney is also working on an image editor. This feature will let you make changes directly within Midjourney, similar to some editing software. You might be able to select areas of an image and have Midjourney regenerate parts of it. More details on this will likely be shared in future updates.

Wrapping Up

The latest Midjourney V7 updates bring exciting improvements and new parameters like `--xp` and `--or`/`--ow`. While these tools are powerful, they often require experimentation to get the perfect result. Keep playing with the settings and exploring the possibilities.

Automating repetitive tasks or large test batches can save you time and effort when exploring new Midjourney features. Tools like the Midjourney Automation Suite from TitanXT are designed to make your workflow smoother.

If you have questions about these updates or want to share your own experiences, feel free to leave a comment below!