DALLE-2 vs Stable Diffusion vs Midjourney

preview_player
Показать описание
The battle of the AI powered text to image producers is here. In this corner we have DALL-E 2, in that corner we have Stable Diffusion and in that corner we have MidJourney and in that corner... ok, well, we only have 3 competitors, so that corner is empty.

We look at all three image powered art tools in action in this video in a variety of different scenarios.

Links
-----------------------------------------------------------------------------------------------------------
-----------------------------------------------------------------------------------------------------------
Рекомендации по теме
Комментарии
Автор

Stable diffusion can be challenging, but you have so much control with many scripts to create animation, morphs, tilings etc.... And it has no censoring like the others.
And you can train it on your own art style and subject like your face.
MidJouney all images have a very... "MidJouney" style.

whatthequack
Автор

stable diffusion all the way! it has evolved so much by the community because open source, it's just beautiful.
i'm using textual inversion to create more concept art for some of my characters. plus no BS censorship.

paulorodriguez
Автор

Stable diffusion is actually really powerful and it didn't get showcased super well here. It does take some experience to know how to write the prompts. I think they tend to understand better when you have a bunch of short, comma separated phrases.

wpelfeta
Автор

The use of commas to break sentences into key phrases is important for the AI. Generally like this:
primary subject, details, details, rendering options, art style
sad woman, beach at sunset, burning sky, photorealistic, shadows.

toddzircher
Автор

The test and testp features in mid journey takes it a notch up . I am yet to use SD, but from the videos I have seen, SD is pretty close and has more room for changing prompts, in painting out painting etc

bilalalam
Автор

The Best Part Is That Regardless Of Which Ones Ultimately Become Ubiquitous In Which Area.
This Text To Image Programs Are Ultimately A Stepping Stone To AI 2D Animation.
One Of The Techniques In Stable Diffusion Is Posing A Model On a 3D Program, Likely Blender, Render It And Use The Render To Guide The AI In To Creating The Correct Pose On The Image.
Once That Program Does Arrive It Will Be The Democratization Of Animation In The Same Way Unity Was For Gaming.
And It Will Probably Be A Blender Addon.

viddarkking
Автор

You can set up midjourney to send you each generated image separately (or just ask it by reacting with envelope)

Drejzer
Автор

Mike its not you fault but you have done this subject a disservice.

Stable diffusion is excellent once you know how to use it properly, it is also going into Photoshop, Gimp, Kritta via plugins so you can Inpaint.
I have seen videos of game artist showing how they use it to greatly speed up their work flows.

These models are already revolutionary, some sites are banning AI generated art.

TheReferrer
Автор

Prompt craft is a thing. The way to get the AI to give expected and good results in a desired style. Respectfully, the way you did it is not ideal to show capabilities of each AI.

isatche
Автор

As someone who has used MidJourney quite a bit, there are image parameters that can be added at the end of a prompt. For example add "--v 4" for the version four of the ai. Seems to make realistic images very well and interpret your prompt much better. And more like "--ar x:y" for a custom aspect ratio.

homecactus
Автор

IMO Midjourney is best for community and hobbyists, it has a lot of options you can tweak, I have the standard subscription because you can generate unlimited images if you are willing to wait a max of 5-10 mins plus the community is usually pretty good at helping you and they do weekly Office Hours talking about anything new coming down the pipeline and taking questions from the community. I haven't used DALL•E 2 much and I would love to use Stable Diffusion locally but my poor GTX 1050 laptop can't run it but I like some of the results. :)

TechieCrow
Автор

I will NEVER use any software that is based inside of Discord.

hellishinc
Автор

To be fair, Dreamstudio Lite is NOT Stable Diffusion. It's a really stripped back version that has no where near the generation options that the full version has.

The_Rising_Ape
Автор

DALL-E Is Doomed Unless They Change Their User Polacy As It Will Gimp The Program In The Medium To Long Term.
Stable Diffusion And Its Myriad Of Children Like Waifu Diffusion Are Free, Open Source, Can Be Specialised To Certain Art Styles And Run Locally.
I See Them As The Best Contender In The Long Run.
I Am Not As Familiar With Midjourney But If They Use Discord And Are Subjected To It's Ever Changing Terms And Conditions It Can Run In To The Same Problem As DALL-E In The Long Run.

viddarkking
Автор

For artistic or concept art Midjourney and Stable diffusion are good at that, but when it comes to generating 3D textures and photorealistic things DALL-E is on another level; the amount of detail and texture it produces in a photo is just astonishing. Use all of them😁

apatsa_basiteni
Автор

Midjourney released V4 recently (still in beta) which kills the competition by a long shot in my opinion.. I ran the same commands as you did in the video and they all came up significantly better.

anestispneuma
Автор

There is also the WOMBO Dream for the fourth corner. It behaves slightly differently with the style choice, but it's an honest competitor (and I like it).

Alche_mist
Автор

I'm working on a boardgame and creating my art with MidJounrey but looking to move to Stability Diffusion. Honestly, I find Midjourney to be way more interesting in terms of results. I'm running SD on my gaming rig. A ryzen 7 3700X, 32GB of Ram and a RTX 3070 locally and it works great.

EricLefebvrePhotography
Автор

Hello, first of all thank you for the video, it’s great content! I would only like to point out something about the prices since you talk about it on the beginning.. midjourney’s $30 usd monthly plan offers a “relax mode” which lets you generate prompts a bit slower without using GPU time.. this way I generated over 9, 000 images with a single monthly payment.
I think it’s very relevant, since its pretty much an “unlimited” plan that the others don’t offer at this time.

joseasse
Автор

You are not looking at the Midjourney final images. The results are incomplete until you've made at least 1 upscale.

Debaser