top of page

Midjourney V6 New Consistent Style Feature — “ Style References” is Here 🎨

Updated: Feb 5

Welcome guys! It’s another round of quick updates from Midjourney. Exciting features are coming soon to Midjourney!

Midjourney introduced an exciting new feature —The “Style References” feature that allows you to replicate a style and maintain the aesthetic easily. Plus, Niji 6 is out and the new” describe” will be arriving next week!

Let’s explore the new features! 🔥

A futuristic guy

🔥 NEW Consistent Styles Feature — “Style References”

Midjourney has just released their first test algorithms for Consistent Style— a highly anticipated new feature called “Style References”.

This new feature allows you to specify one or more reference images that define the aesthetic you want Midjourney to maintain across your generated images.

What is “Style References”

Style Reference works similarly to image prompts where you give a URL to one or more images that ‘describe’ the consistent style you want to work over.

Highlights of Style References

  • Functionality: Similar to image prompts

  • Usage: Type --sref after your text prompt, followed by URLs to one or more style reference images like this: --sref urlA urlB urlC

  • Format: /imagine text prompt --sref url

  • Style Matching: The model looks at image URLs as ‘style references’ and tries to generate something that visually ‘matches’ their aesthetics.

  • Goal: Create output matching the provided aesthetics

  • Compatibility: Works for both V6 and Niji V6 ( Not work with V5 etc)


Example of “Style References”🎨

I took the lion image on the left side as a reference image and replicate replicate the style and generate the fox on the right side with the new “Style References” feature.

A simple step-to-step guide will be

Step 1: Pick Your Reference Image

Use an image that you generated with Midjourney or pick one from the internet

Step 2: Upload an Image or Copy a URL

Upload the images to Discord.

A holographic lion

Step 3: Type Your Prompt

Type --sref after your text prompt, followed by URLs to a style reference image like this: --sref ur

An example will be like this /imagine prompt: Your Text Prompt --sref urlA

With the New “Style References” feature, I can replicate the style easily

/imagine prompt: A fox --sref url
Replicate a style with “Style References ”. Image reference (left) and new image (right)
Replicate a style with “Style References ”. Image reference (left) and new image (right)

Character Consistency 🎨

Meanwhile, the consistent character system is currently in development, with upcoming plans for:

  • It will function similar to Style References with the --crefargument

  • Initial Trial: Simpler versions to be attempted first with potential future changes and version management


Niji V6 Alpha ✨

As some of you may notice from my previous update, another exciting news this week is the launch of Niji V6 Alpha!

Niji V6 has a different aesthetic and improved prompt understanding! The most exciting thing is that it can now generate simple text just like Midjourney V6!

Turn On Niji 6

To generate with niji 6, use the suffix--niji 6 or go to /settings in the niji bot, and turn on Niji V6 [Alpha].

Niji 6 alpha

Highlights of Niji 6

  • Improved Prompt Understanding — Try longer, more difficult prompts

  • Simple Text Generation: Write simple text by putting it in quotations.

  • Raw Style — If you don’t like anime use --niji 6 --style raw

  • 📅 The full launch will be in February — Expect to see Vary (Region), Pan and Zoom soon

Here are a few examples created with the New Niji 6

As you can see, we can now generate simple text and have more control over the details of the images.

To learn more about Niji 6, here is an in-depth article where I share 4 Tips to create stunning anime art with the new version!


Midjourney Website 💻

Eligibility Expansion:

  • The MidJourney Alpha website is now accessible to users with over 2.5K/1K generations.

  • Check how many images you’ve made with /info

  • You can access the site here:

Outpainting Features on Web:

  • Introduction of outpainting features “Zoom” and “Pan” on the web platform

Tranlucent computer surrounded by beautiful flowers

👉🏻 More Upcoming Plans

A new /describe🎨

The MidJourney team is working on an entirely brand-new /describe feature. The goal is to make the new feature able to duplicate images more accurately.

  • The new /describe will be more powerful and accurate than before.

  • It is currently under testing

  • 📅 Estimated Release: Next Week

🎨 New Style Tuner

A New Style Tuner will be released in February 📅

The upcoming version of Style Tuner will differ significantly from the current one. A New Tuner optimized specifically for V6 be created, as the V5 Style Tuner won’t be compatible.

Midjourney V7

The MidJourney team is prioritizing the launch of MidJourney V7, set to be released in the next three months.


👉🏻Passive Income with AI art: Master Midjourney 🎨

Don’t Get Left Behind in the AI Art Gold Rush of 2024! Midjourney is exploding. Early adopters are raking in big bucks selling AI art online.

This is your ticket to passive income with AI art.💰- “Master Midjourney 2024: Beginner to Pro” digital guide.

It comes with:

  • 🎨 1,000+ money-making prompts for stunning, sellable art

  • 💡 Advanced Midjourney prompt writing, structure and techniques

  • 🖼 Specialized commercial guidance for graphic design, branding, logo, UI, photography, mockup and more

  • 🚀 Insider tips to maximize profits

  • 💰 AI Arts Monetisation Strategies

  • 💼 Top AI tools to edit Midjourney creations

Master Midjourney 2024

📚 Continuous education

This guide is designed for continuous education, new tutorials are added regularly so you stay ahead. With lifetime access for a single purchase, it’s the gift that keeps giving.

I’ve poured my heart into creating this guide. I hope you enjoy it ❤️

3 views0 comments


bottom of page