Midjourney Moodboards is here!

Now you can throw a bunch of images and train your own model with Midjourney.

Details ↓ Multiple Personalization Profiles
You can now have multiple profiles for personalizing your experience. No need to stick with just one!

Faster Setup
Setting up personalization is now 5x faster! You’ll get your model tailored to your style in no time.

Moodboards for Personalization
You can now personalize your models by uploading moodboards. Just upload images, and the model will use them for inspiration

Quick Convergence
Personalization profiles now converge more quickly. You’ll need just 40 ratings to get started, and the profile will be stable by 200 ratings.

Diverse Remixing
As you add more diverse images to your moodboards, the model will begin to remix them in more complex and creative ways.

Custom Profile Names & Defaults
You can name your profiles and even set one or more as the default for easier switching between styles.

Track All Images
You can now see all images created with a particular profile. Perfect for tracking progress and consistency.

Why This Is Important:
Whether you’re working on multiple projects or collaborating with a team, these features make it easier to manage and personalize your creative journey.

Midjourney v6.1 is here!

There are many improvements.

Here's what's new: New Personalization Model
Introducing an enhanced personalization model with improved nuance, surprise, and accuracy. Now, your code changes as you rate. You can still use your old codes anytime.

Improved Details:
More accurate and detailed small image features (eyes, small faces, distant hands, etc.)

Faster generations
Approximately 25% faster for standard image jobs.

Better Coherent:
Enhancing arms, legs, hands, bodies, plants, animals, and more

--q 2 is back!
A new --q 2 mode which takes 25% longer to (sometimes) add more texture at the cost of reduced image coherence

Improved Quality:
Fewer pixel artifacts and enhanced textures, including skin.

New Upscale Models
Enhanced 2x upscalers with superior image and texture quality

Better Text Rendering
More accurate text when using "quotations" in prompts.

Overall:
v6.1 is more beautiful than ever :)

Google just introduced Imagen 3.

Their latest text-to-image Gen AI model.

Let's compare it with Midjourney v6 ↓ Prompt: Three women stand together laughing, with one woman slightly out of focus in the foreground. The sun is setting behind the women, creating a lens flare and a warm glow that highlights their hair and creates a bokeh effect in the background. The photography style is candid and captures a genuine moment of connection and happiness between friends. The warm light of golden hour lends a nostalgic and intimate feel to the image

Google's Sora is here: Veo

It can generate 60 sec 1080p videos!

8 examples:
Prompt: Many spotted jellyfish pulsating under water. Their bodies are transparent and glowing in deep ocean.

How to use the new Midjourney Web?

I see many of you're excited for the new website, but you might not know how to use some hidden features.

From my past posts, I've identified key topics. I'll simplify them and show you through short videos.

Get ready; we have a lot to cover! Please note that, Midjourney web is still in the alpha testing phase, so what I show here may change later on.

If you've generated 100+ images, you can try it out:

alpha (dot) midjourney (dot) com

First *real* AI website builder ever!

Introducing: Dora AI Beta

🧵 Link and more in the thread: @DoraTool uses natural language to create customized websites, including layouts, images, text, and logos.

While it's not flawless yet, you can edit the generated site as needed because it's also a no-code editor.

You can try it right away:
a.dora.run/iw8znb4sag09

This new 'Character Reference' is insane!

Now, character consistency is possible in Midjourney.

🧵 Here is how and examples: You can get this guide as a PDF here:
doganuraldesign.lemonsqueezy.com/checkout/buy/1…

We're stepping into a new reality (!)

Introducing EMO: a groundbreaking AI that creates expressive portrait videos from just an image and audio!

🧵 More in the thread: EMO generates expressive portrait videos from just an image & audio! It captures subtle facial expressions & head movements, creating lifelike talking & singing videos.

I'm compiling the entire X algorithm repo to upload it to Gemini 1.5.

What would you like me to ask? Compiling is done!

Token Count: 310489

The Sora team has started sharing new videos.

Some of them are quite interesting.

🧵 Here are 10 examples: 1/10

https://twitter.com/630587365/status/1758966425526378766

OpenAI just dropped the bomb on video generation.

But how do they compare to their closest rival?

OpenAI: Sora vs Runway: Gen2 Prompt: A Chinese Lunar New Year celebration video with Chinese Dragon

Everyone needs this:

Ultimate Cheat Sheet: Midjourney Parameters

🧵 More in the thread Aspect Ratio: --ar or --aspect
Chaos: --c or --chaos
Weird: --w or --weird

The new 'Style References' is mind-blowing!

I tried to transfer the styles of some movies.

🧵 Here is how and the results: What is it?

Midjourney Style References are here!

Now you can generate more consistent styles.

🧵 Here is how: What is it?

Marbling Landscapes

🔓 Unlocked with Midjourney v6

Prompt and examples in the thread: Prompt:
aerial photo of a landscape, {location}, in the style of ebru, detailed imagery, organic formations, multicolored --ar 4:5 --style raw --v 6

Marbling Portraits

🔓 Unlocked with Midjourney v6

Prompt and examples in the thread: Prompt:
portrait of a humanoid in marbling art, {color combination}, emotive facial expressions, realistic liquid motion, in the style of ebru, organic and fluid, --ar 4:5 --style raw --v 6

Unbelievable 32x zoom

NanoLand: Day 02
“She’s on vacation”

🔊 Sound on

Prompts and settings are in the thread 👇 Enhanced with → @Magnific_AI

Meta's AI image generator has just been released.

I have tried the prompts I used with Midjourney before.

More in the thread: Universal Prompt:
Minimalistic portrait of a 70s hippie woman at beach, expressionless, headband, 70s hairstyle, 70s fashion, striped shirt, sun rays, light teal and amber, soft hues, Cinestill 50D

New FREE AI enhancer has just been released.

So, I had to do it...

Magnific AI vs Krea AI (Free)

More in the thread: Landscapes and architecture

Google just entered the game!
If you can't find the image you're looking for, it allows you to create something similar

Adobe's Firefly Image 2 just released.

I have tried the prompts I used with Midjourney and DALL-E 3 before.

More in the thread: Universal Prompt:
Minimalistic portrait of a 70s hippie woman at beach, expressionless, headband, 70s hairstyle, 70s fashion, striped shirt, sunrays, light teal and amber, soft hues, Cinestill 50D