Why Midjourney 4 Has Just Gone Viral

A stuffed pikachu has gone viral, wow!

Two. No, I spent hours and hours and hours in midjourney four. The results are quite frightening. So you might be wondering what’s new in this mid journey. It’s an alpha version.

The full version has not been released. They have been building this for nine months with help from yours truly, you and you and everybody who has used the Discord server. They’ve been taking all this information and seeing what humans actually really like. There is way more information and knowledge in this model, which means there’s more creatures, people’s, places and way more intricate detail within those images than we have seen before. They’ve been collaborating, that all, with new AI architecture and have built midjournie for one of the main features is that it handles multiple subjects as well as multiple people.

You can put multiple people in a scene interacting with each other, which is insane and it works really well. The only issue is that we still have that nightmarish hand issue where the hands all form into each other, emerge and have six or seven fingers. They’re probably working on that and I can imagine when they release the final version that will be fixed and again with the objects as well, you can put multiple different objects in a scene and use prompt layering to direct each object. It also supports chaos, so you can set it from zero to 100. So this is the alpha version.

So this is the introductory version. And as I was saying, the resolution and quality is not as of the previous versions because they’re holding that back. For the final version. You cannot currently do aspect ratios. Again, they’re just working and tweaking things out of that.

So you just stuck with the same aspect ratio. And the four quads you can upscale slightly for a better version. You’re not going to get highquality different aspect ratio out of it. Basically, you need to expect quirks and inconsistency, this being an alpha version. So if there’s any errors and things going wrong, hopefully that will be fixed.

In version four. And this is straight from the horse’s mouth, this is quoting the journey. They’ve said version four isn’t the final step, but our first step and we hope you all feel it at the new beginning of something deep and unfathomable. They are moving forward to the full launch, whenever that will be. There is no date set for that, but for now we can play around with this alpha version.

And it’s looking really promising. So come with me, I’ll show you some things I created. Let’s have a look at these. So this is what I found really interesting. So if you sell on Etsy or Ebay or prints or whatever, you’re probably used to going to shutterstock or wherever, get Images, to name a few, to get your mockup images.

Now I actually create mockup images in blender, but this here, I’ve just typed in a a four white frame on wool focus on frame in the Stylish home. Realistic, some of these images, they’ve given me the artwork in the frame, which is not a problem. I can remove that very easily in postproduction. Now, this test was quite interesting. So prior to version four, I had some big issues with motion action shots.

It was adding extra limbs and I was having to remove them with the Remix feature, and it was just all a bit cumbersome. Now they seem to have nailed the action shots. And I loved these. I wanted to see what the photo realism was like and I did the plus size model photoshoot, and look how beautiful these images are. If I just click on I mean, that looks almost real.

That’s insane. To be honest, I don’t know if I would be able to tell the difference between this image was in a magazine. I wouldn’t be able to tell you that it was created with AI. That’s insane animal photography. So this is using the multiple subject prompt.

So I’ve put a Dalmatian dog, a golden retriever and a Jack Russell sat together. Pet Photography Ultra HD, focus on dog blow edges, super detailed, et cetera, et cetera. And yeah, I mean, this one is gifts. Give me one dogmation for some reason. But that’s really I actually spelled at Russell wrong.

That’s why I was just trying to put another dog in, because my spellings are atrocious. Who doesn’t want to dress their dog up as baby yoda? You can put outfits on your dog with the algorithm. Moving on to a kind of octane blender render. Blender render.

I just went for an aerial character. Funny octane render, which is a render. So when my journey eventually comes out with their 3D programme, I suspect these will be built with polys, which is insane, and I’m looking forward to seeing that. But, yeah, it’s generated something that you would expect to be on the front of 3D Arts, if that’s still around, and you’re old like me. Going back to the photography, I wanted to just try out like a little girl dressed up as a fairy.

Photography Ultra HD and this is what it’s given me. And then again, you can upload a picture of your child. I wouldn’t be uploading a picture of my child to the Internet. There you go. And then add this style on to them.

This is a 3D Isometric withered room within Blender. And again, it’s served. It’s given me what I’ve asked for, frighteningly fast and accurate. And again, you can change the details in this image with the Remix feature. I then changed it to be a 3D Isometric dungeon room Blender.

And there you go.

I’m still on the 3D theme. I went 3D low poly game hut Blender. Again, I’m using a really short to the point prompts, and they said that you can add more detail and multilayering of prompts, which is really cool. But, yeah, it’s giving me like Pinterest 3D game art results, which is crazy. I don’t think many industries are safe from this.

I know Mark Holtz had said that he’s working on a 3D programme. With another click of a button, you’d be able to get the polymesh for these game assets, which is insane. Yeah. Oil painting of blackbird bold brush strokes. And it’s just again, I wouldn’t be able to tell you that AI painted that.

I just did a few tests on the bold brushstrokes as well, and getting that stylization as well. But what will be cool is that you can upload a picture of your dog or yourself and add this style to your image, and it’s a very satisfactory look to the image. And I know my journey had been working on looking at what people were actually creating in discord and what pictures they liked and what pictures they didn’t. And this taught the algorithm what was pleasing to the human eye. And I think they really have got this in Mid journey.

I then moved on to watercolour, like loose watercolour. So this is Watercolour zebra, loose brushstrokes watercolour. Again, because I wanted to emphasise the fact that it was watercolour, because sometimes if you don’t put it in multiple times, it won’t create that loose, watery style. I mean, look at the top right one. That’s insanely beautiful.

This is really good. This is better than I could paint, which is worrying. This isn’t at all. This is a replacement. So I wanted to see a bit more detail.

So I put in a felted mouse holding a flower. Cute. Stupid detailed. My journey is serving us, guys, on that handmade theme. Again, I’m looking at intricate detailed here.

I did an embroidery styled needlework of deer and fox and this is what it gave me. Look at the details. You can see the actual embroidery. Again, with the watercolour style. I tried to recreate, like, Etsy birthday card sort of thing in watercolour style.

And as you can see, watercolour and ink Georgia with a party hat on. And baby, that’s what it gave me. I then tried a Beatrice Potter like, I guess like a child book style artwork, a fog dressing clothes by Beatrix Potter, watercolour style. And bam, it gave me that. And look at the detail of the clothes.

It’s amazing. Again, midjourney is a better artist than I am. And then I looked at, again, a trending image on Etsy and recreated that in midjourney. And it’s far removed enough away from that image on Etsy that I probably copyright this and resell. So that’s a black and white photo of an elephant in a buffer, but photorealistic.

Yes. Thank you very much, guys. And on that note, I have 500 subscribers. I don’t know why you guys are subscribing, but I like you very much for doing that. But the reason why I love doing this channel is that it can help open up the AI community to those that are unaware of it or do not have much knowledge of digital art and AI art.

And that’s what I’m here for. I actually base my my career on learning from YouTube videos. I know. But I managed to forge a career somehow in it. And that will do it for now.


Join The Prompt Muse Gang!

Get the latest A.I News, Reviews, and tutorials, hand-picked by A.I depending on your preference and sent directly to you.

A newsletter featuring two pictures of a man and a woman highlighting 'prompt news vs muse'.

*We promise not to spam, sell or do any other naughty things with your details – simply keep you updated when we launch new awesome tutorials and news.