Midjourney Version 6 - IS AMAZING!!!

Summary

The video discusses the new version 6 release of Midjourney AI and demonstrates how it has improved image generation capabilities. The host shares excitement about using Midjourney almost daily for expressive and stock-like images. Some key improvements in version 6 highlighted: - More accuracy for longer prompts - Increased coherence and knowledge, allowing more complex scenes - Less need for 'chunky' prompting conventions used in v5 - Ability to generate readable text Examples demonstrate v6 improvements: - Simpsons x Beauty and the Beast mashup retains styles well - Santa Claus image renders tongue realistically - 'Hello World' text renders clearly several times - Dragon battling warrior scene has excellent composition - Cat reading book image has cohesive style and pose - Sci-fi cockpit has functional, coherent elements - Giant robot over city street has effective scale - Old arcade scene captures lighting and mood well - Warrior against monster has strong color contrast Stock photos generated: - Food photo with appetizing realism - Office worker with skyline background looks authentic While giant woman walking through village is not perfect, it shows improved capabilities for complex prompts. Key takeaways: - V6 has significantly improved image coherence, accuracy, and complexity handling - Text rendering works fairly well now - Stock photos are generated very realistically - Will need to refine prompting approach compared to v5 - Still an alpha but available to try now and provide feedback Overall, v6 demonstrates major strides for Midjourney in generating accurate, high-quality images for a wide range of prompts and use cases.

Takeaways

• Midjourney version 6 brings major improvements in image generation quality and accuracy
• It has better coherence, composition, and artistic direction in complex images
• Text rendering ability is improved in v6
• Longer prompts work better now
• V6 images look more realistic, like actual photos and stock images
• Reprompting strategies from v5 may not work as well in v6
• Describe feature will be upgraded to v6 later
• Additional features like panum, very wide etc are coming soon
• V6 handles prompts for complex scenes like spaceship interiors better
• Overal v6 allows creating images closer to the intended prompt

Q & A

What are some of the major changes in Midjourney v6?

Better image quality, more accurate prompt following, improved text rendering, better results for long prompts, more realistic image generation.

How is artistic direction and composition better in v6 images?

V6 makes smarter decisions about lighting, contrast, saturation etc that enhance the overall artistic quality of images.

Why do old v5 reprompting strategies not work as well now?

V6 has been significantly upgraded so previous tricks like adding 'award winning photo realistic' don't improve quality much now.

What additional features are planned for future v6 updates?

Panum, very wide, upgraded describe v6 etc. Some announced features like text drawing are already partly implemented.

How does v6 better handle complex scenes like spaceship interiors?

It can now render complex functional scenes better even when few example images exist, thanks to improved model knowledge.

What does the improved text rendering ability allow?

V6 can now accurately generate written text like signs and labels within an image based on prompt.

Why do v6 images look more like real stock photos?

Enhanced coherence and realism in rendering various elements like food, office scenes results in more authentic looking images.

How can describe feature help in future?

Describe v6 will allow generating better prompts automatically that are tuned to capabilities of v6 model.

Why is relearning prompting strategies needed for v6?

The significant upgrades mean v6 works quite differently from v5 so previous prompting tricks don't apply well.

How does v6 allow getting images closer to prompt?

Increased accuracy, model knowledge and improved handling of artistic direction results in images that match prompts better.

Outlines

00:00

Introduction and Overview

Introduces that a new version of Midjourney is out and discusses some of the overall changes and improvements. Shares example images and prompts to showcase the capabilities. Also provides a personal update on the video creator's travels in Thailand.

05:02

Image Examples and Prompts

Shows specific image examples created with Midjourney v6, including a Beauty and the Beast / Simpsons mashup, a cartoon Santa Claus, and text rendered as images. Discusses how v6 handles complex scenes better like dragons, landscapes, and sci-fi spaceships. Also tests anime cat drawings, abandoned arcades, giant robots, and more.

10:02

Stock Photo Examples

Tests Midjourney v6's ability to create realistic stock images using food photography and office scenes as examples. Shows how it can generate high quality, authentic looking images that could be used in ad campaigns and commercial purposes.

Creating Gigantic Figures

Attempts to create images of gigantic figures integrated into landscapes, which is still challenging. Shares an example prompt and resulting silhouette of a 20-story tall woman walking through a village.

Announcement and Alpha Test Details

Summarizes details from the Midjourney v6 announcement, including improved accuracy for long prompts, better coherence and knowledge, and not needing certain prompting phrases used in v5. Mentions new features and capabilities that are still in development.

Keywords

💡midjourney

Midjourney is an AI image generation system that creates images from text prompts. The video discusses improvements and new features in version 6 of midjourney. It shows examples of images created with version 6 and analyzes how the quality and capabilities have advanced.

💡prompting

Prompting refers to the text prompts or instructions that are input to midjourney to tell it what kind of image to generate. The video examines how prompting strategies may need to change for version 6, with less dependence on certain prompting conventions used in version 5.

💡coherence

Coherence refers to how logically coherent, consistent, and meaningful an AI-generated image is. The video praises version 6 for improved coherence in complex scenes with multiple elements interacting logically.

💡artistic direction

Artistic direction refers to visual design choices that enhance the aesthetics, emotion, focus etc. of an image. The video analyzes strong artistic direction in several version 6 images related to lighting, contrast, saturation etc.

💡composition

Composition describes how elements are arranged within an image. The video frequently compliments version 6 images for excellent and meaningful compositions that align with the desired scene.

💡depth of field

Depth of field refers to the area of an image that is in focus. The video notes clever uses of depth of field blur in version 6 images to guide the viewer's attention.

💡silhouette

A silhouette refers to the outline or general shape of something, with interior details in shadow. The video points out the use of lighting and contrast to create strong silhouettes that help subjects stand out.

💡stock images

Stock images are reusable generic images that can be used across different contexts. The video examines version 6's ability to generate believable stock photos for areas lacking good actual stock images.

💡describe

The describe feature summarizes an AI-generated image into a text description. The video notes that a version 6 update to describe will help refine prompts to achieve a user's desired type of image.

💡capabilities

Capabilities refer to the spectrum of visual concepts midjourney can depict and the quality with which it can render them. The video discusses expanded capabilities in version 6 to handle more complex scenes and elements.

Highlights

Midjourney version 6 brings amazing new changes for more expressive and stock-like images.

New version has better accuracy for longer prompts created with GPT-3.

The AI model has more coherence and knowledge to create complex scenes.

Version 6 is trained on more diverse data to generate a wider variety of images.

Significant difference from version 5, will need to relearn prompting techniques.

No longer needs text prompts like 'award winning photo realistic'.

Now has minor text drawing capabilities to create text in images.

Additional styles like weird, subtle, remix, describe now work with version 6.

Describe version 6 coming soon to generate better prompts from images.

New version creates high quality, believable stock photos.

Can easily create authentic-looking office scenes.

Still challenging for giant characters, but quality improving.

Artistic decisions improved - contrast, lighting, color harmony.

Better functionality, coherence in complex spaceship interiors.

Exciting advancements, relearning prompts needed to benefit.

Transcripts

midney version 6 is out and brings some

amazing new changes hello my friends how

are you doing I'm actually using moury

almost on a daily basis especially if I

want to have very expressive images or

stock likee images so these new changes

are actually really good I'm going to

show you some example images and prompts

and talk about how they changed and why

it is so important to have these changes

also shout out to robar for helping me

with this video also here's a little

update for my Thailand St it's just very

beautiful here I'm seeing fireworks

these amazing dance performances that

are in the shopping centers I have spent

some time with the monk talking about

Buddhist philosophy and saw this

gigantic Buddha statue shout out to

Jurgen thank you for organizing that day

for me of course making some new friends

and meeting the Thailand stable

diffusion Community all of that is

really nice let's get started now it's

time to look at some images and the

prompts here so this is a prompt from

robar Beauty and the Beast and the style

of The Simpsons it's creating that in a

pretty good way it actually looks like a

scene from a Simpson comic so that is

very nice easily understood now of

course you will see when you look a

little bit closer there's a little bit

of a problem with the interaction of the

hands and the arms but the style itself

is pretty nice here's another example

image from robar Santa Claus Superstar

showing tongue cartoon hyper realistic

now there's a little bit of a version

five prompting in here but the

interesting thing is the tongue sticking

out and this is something that was

pretty hard to do in version five

especially in a way that looks good I

would say here it looks really good now

here's an example that I created with

the text hello world and in three out of

four Images this text was actually

created correctly and I also like here

it's not just creating the text but also

you can see there is some depth of field

blur on the text itself so the effect

here is pretty pretty amazing here you

can see the prompt that is pretty simple

a photo of the text hello world written

with a marker on a sticky note and this

is exactly what we got here's the next

scene that I had to test out we have

here a very large dragon and then a

smaller character a warrior standing on

the left side in the ruins of a castle

and I have to say this turned out really

well this is a first Roll image and the

composition is very nice but also

followed my prompt very very nicely so

here you can see the prompt I'm using a

Burning Castle ruin on the left is a man

in Warrior armor looking small in the

image he's looking up to a large red

dragon who is looking down at him with

anger and I have to say that this

actually is completely there even the

interaction with the character the whole

style is there the fire is there the

colors work very nice and a lot of this

also has very nice artistic Direction in

the sense of what is more or less

saturated what is brighter and darker

where we have high and low contrast so

an example here would be that the

warrior has as a background lighter

color so you can actually see his

silhouette and even the sword he's

holding in his hand very good artistic

decision here from M Journey as you

probably know I love cats and I love

lowf music so of course I love lowf

music wallpapers here I wanted to have a

humanoid cat sitting in the sunset

reading a book this kind of cozy

scenario and this turned out really

amazing now this is also a first rle and

I'm really really happy with everything

here the art style the composition the

humanoid structure of the cat how the

cat is actually reading and holding the

book all of that is really beautiful

here we have the prompt for that Co

drawing in anime style of a cat with a

human body reading a book in Sunset

light this is not using Nichi this is

using mid Journey version six here's

another scene that can be quite complex

for Mid journey I have to say I think it

has improved a lot now here we have the

cockpit of a Sci-Fi spacecraft and this

is interesting because of course this is

not something where there is thousands

of images online that actually show you

how stuff like that looks so in the P

mid Journey had a lot of problems of

creating these Interiors of spaceships

because there's just not many examples

there so I feel like here we have a very

nice structure very nice composition we

have a lot of functionality in the

different elements in here and also the

character is sitting correctly on a

chair that actually makes sense and he

is sitting there actually doing work so

all of that has a lot of coherence and

all of that has a lot of functionality

in there here we have the prompt in this

case I have used chat GPT to create a

longer prompt it reads the spaceship's

command center is a hive of activity

with holographic displays the captain

overseas operations from a central chair

surrounded by a panoramic view of space

through a transparent viport crew

members in futuristic uniforms operate

control panels the air is charged with

urgency as the crew coordinates their

efforts and the soft hum of the

spaceship systems underscores their

focus next we have your giant MEC robots

standing in a city street also very nice

again I love the dynamic between the

different elements how he's standing

there how the light is working in the

same way on him and on the surrounding

buildings and then as a contrasting

elements we have down there at the

ground very small people that are

running around in the streets again you

can see here the very nice decision that

the Silhouettes of the people on the

ground have as a background lighter FX

so that we can actually see them very

clearly I love also not just the

composition but also the camera angle

looking up at that large robot and

giving us very nice reference point on

how big this robot actually is here we

have the prompt very simple in this case

movie scene of a giant robot standing in

a city another thing I really like is

linal space and here we have an old

arcade I wanted to have some green light

here we see some dirt on the ground some

of the machines are still turned on so

all of that is a pretty cool very

interesting scene love the lighting love

the atmosphere in this again here we

have a little bit of a longer prompt

created with Chad gbt it reads photo in

Gray and green tones of the dimly lit

space of an abandoned arcade the

flickering glow of the malfunction

neolites cast an eerie Ambience the once

vibrant arcade machines stand silent and

lifeless they greens Frozen in Time dust

particles dance in the air caught in the

soft beams of faded pastel colored

lights the air is thick with a sense of

nostalgia and abandonment as the ghostly

Echoes of long-forgotten game sounds

linger in The Emptiness and then I have

to say again a lot of that is found here

in this image with the dirt on the

ground with the broken ceiling it has

this kind of linal eeriness in there so

that's pretty cool here's another

example of a dynamic between a very

large monster and a warrior standing in

front of it again I really like the

composition here with the cave with the

water flowing down and again I can't

stress this enough the nice decisions of

the color here most of that is blue and

orange and we have most of the light and

the fire where the hero is standing and

it is really bringing out the silhouette

of the heroes standing there and the

dynamic to the monster because also

because of that light reflecting on the

face we can see that the monster is

looking down at the hero so there's so

much artistic nice decision in here and

this is really what moury is very very

good at again we have a prompt from Chad

GPT at the entrance of a dark cavernous

abis a stoic Warrior faces a colossal

shadowy Titan emerging from the depths

the air is thick with tension as the

warrior based in the e glow of the

enchanted runes confronts the Monstrous

entity the clash between the two Titans

Echoes through the cavern each strike

resonating with raw power as the warrior

battles to prevent the malevolent force

from escaping into the world above next

I want to look with you at some stock

images and I personally think that this

is also a Hallmark of good AI image

generation how well do these look as

actual stock images that you could use

for a campaign how much do you believe

the actual image and here I have to say

the results are really good the food

looks really nice and tasty and like

actual food in there the salad the

tomatoes the cheese the meat even the

sauce in between the salad and the bun

looks like actual sauce all of that

really good the composition is very nice

with the depth of Fieldy have a nice

focus on the food itself so all of that

looks like an actual product photo

really really good here we have the

prompt you can see it's very simple

stock photo of a juicy burger with

tomatoes and salads here's another

interesting scene and this is one of the

things I love to test these are office

pictures because you will find that if

you go to stock Pines and what to have

any kind of office or business related

image they mostly look really terrible

really staged really not authentic at

all so here I want to have a guy who

who's working late at the office with

the illuminated City in the background

and mney has delivered on the first role

now this is the future of stock images

because instead of searching for hours

for good image you can just create it

like that here we have my prompt man

working late in office sitting in front

of a computer at night few of the city

skyscrapers behind him through the large

glass w windows and look at how

everything is in here and everything

looks really good and here's the last

image that I want to show you today this

is kind of a challenge for you because I

found that creating a gigantic woman

walking through a village or a city is

something that is still very difficult

to create with mid Journey especially in

a way where the woman is not a statue

but actually a woman walking around now

here it is a silhouette with no deta

details and it looks a little bit more

like a monster but still I have to say

in this case MIT has done a really good

job The Prompt here reads gigantic woman

who is 20 stories tall walking through a

village and yes I see that I have a

little bit of a typo here in through

let's have a bit of a look at the

announcement and of course you can set

it up in the settings to use version 6

but you can also use minus minus V6 to

call it inside of the prompt without

changing it in the settings now they

talk here a lot about more accuracy also

for longer prompts which is very good

when you create prompts with jat GPT and

also it has more coherence and model

knowledge in there which is also good

because then it can create more complex

scenes and just understand better what

you mean and the model is trained on

more different kind of things you want

to create and the SC we're going to see

that in a second now something they say

here that's pretty interesting is that

this is a significant difference to

version five and they say you will need

to relearn how to prompt so things like

aart winning photo realistic 4K 8K they

call it chunk and that is no longer

needed for the images not quite sure

100% you don't need that but it actually

is creating really good results even

without these words in there another

thing to point out here is that it

actually has a minor text drawing

ability so it can create text and in my

test actually created a text pretty good

in most of the rendered images and right

now this is an alpha test this is not

the final version of version 6 but it is

out in the sense that everybody can use

it so you can see here that all of these

different features like different ratios

chos weird tille styz styz raw very

subtile very strong remix blend describe

and so on are also available for version

six although for the describe version

says just the version five is used here

and then there is some other stuff

that's not working yet that that's

coming soon that is for example panum

very Reach In Tune and then also

describe version 6 describe version six

will be very very useful because then

you can actually have prompts as they

are meant to be and you can just create

prompts based on images tuned to the use

in mid Journey personally I'm very

Amazed by these changes because they

help me out a lot in generating the

images I need let me know in the

comments what you think about that

thanks for watching and leave a like if

you enjoyed this video bye oh you're

still here so uh This is the End screen

there's other stuff you can watch like

this or that really cool and yeah I hope

I see you soon uh leave a like if you

haven't yet and well um yeah