Midjourney V6 vs Leonardo AI Alchemy- Image Generator Showdown!

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
Midjourney just dropped its latest  game-changer, Midjourney v6! Today,   we're putting it head-to-head with its  biggest rival, Leonardo AI Alchemy. When Leonardo AI Alchemy burst onto the scene,  it made a name for itself with a strong focus on   realism and some nifty features like turning still  images into dynamic wonders. On the flip side,   Midjourney v6 is all about crafting  detailed, dreamy outputs and rocking   a killer art style in its animations. So,  let's find out if the hype holds true. We’re here to help you figure out which  one suits your project best – whether   it's text to image, image to image,  keeping those characters consistent,   or even cooking up some sweet videos.  Let's get this showdown started! Starting with text to image, let's dive  into Leonardo AI Alchemy. Just a heads up,   you need to subscribe to one of their  plans to use it. No pressure, though—I   suggest considering it after watching  this video. Stick around till the end,   and don't forget to hit that subscribe  button for more AI tech goodies. Now, let's hop over to Leonardo AI and give image  generation a shot. Don't forget to turn on Alchemy   V2 on the left side. Leonardo Diffusion XL  excels at artistic styles, Leonardo Vision   XL prioritizes photorealism, and AlbedoBase XL  focuses on detailed character creation. Today,   we're rolling with the AlbedoBase XL. We're going to use the same prompt for   both platforms and here's what I came up with. “A pixel art animation of two baristas crafting   the perfect latte. One barista is making a  steam swirl, another is wearing a colorful   shirt and frothing milk for her latte,  and there are cute pixelated coffee beans   dancing in the background. With a backdrop of  ‘Coffee Please!’” Hit that generate button.  Take a look! These pictures look  really nice. The visuals are awesome,   in my opinion. Even though it didn't quite  nail the background thing, the overall result   is still impressive. You can also use Live  Canvas Leonardo AI and the Leonardo AI Canvas   Editor if you want to tweak your images even more. Now, let's check if Midjourney can do even better.  To use Midjourney v6, all we're going to do is  add a double hyphen and v6 to each prompt. So,   let's type slash imagine, then copy-paste the  prompt we used for Leonardo Alchemy earlier   and add -- v6 at the end. If you  want to include text messages in the image,   make sure to put quotation marks. Now,  let's hit generate and see what we get.  The results from Midjourney are just as good  as Leonardo's, and it nailed the backdrop   this time. It looks like it followed  our prompt perfectly. I've got to say,   it outdid Leonardo's earlier results. Now, let's dive into experimenting with   image to image. We're taking this black and  white image and turning it into something   colorful and lively. With Leonardo AI image  to image, we'll start by uploading it.  For our prompt, let's type something like “Paint  this picture in a bold style of colorful pop art,   each member bursting with personality  and energy.” Switch the fine-tuned   model to Leonardo Diffusion XL, lower the  image strength to 0.20, and hit generate.  Once again, the results didn't disappoint.  It's vibrant, lively, and full of color.  Now, onto Midjourney v6. Let's begin by  uploading our photo. Once it's uploaded,   copy the image URL and add our prompt,  including the double hyphen v6 at the end.   You can also incorporate your Midjourney  Style Tuner if you want specific styles.  While Midjourney did well earlier, it didn't quite  bring its A-game this time. I mean, look at that.  So far, Leonardo Alchemy has been decent,  but let's give it another shot to see if   it can generate consistent characters. To  start, when crafting consistent characters,   the key is to first find the perfect starting  image. We'll use a detailed prompt to describe   our character – and I’m going for something  different like “Generate an image of James,   a realistic green lizard-human guy, 18 years  old. But most of his features are human-like. His   eyes are curious and brown, and he's got a cool  jawline. When he smiles, it's friendly, and his   nose is uniquely shaped. Plus, he wears a stylish  jacket, making him a chill humanoid lizard.”  Now, let's generate at least ten images and  save them for later use in our model. Moving on,   we'll create our model using the  description we provided earlier and   name it James. With our model in place, it's  time to generate some consistent characters.  So, we'll generate James two times in  different situations and positions to test   Leonardo Alchemy's consistency. For example,  “Create an image of James, performing at his   own concert. Picture him in a cool t-shirt,  surrounded by fans, with vibrant stage lights.”  The results are okay, not bad but not perfect.  The images don't exactly match our idea of James.   Let's try placing him in another scenario, like  “Generate an image of James sitting on a bench,   hand extended, with a bird perched on  it and a bright smile on his face.”  Unfortunately, the results don't look great  – there are some awkwardly placed parts,   and the facial features don't  quite capture James' likeness.  Let's see if Midjourney can step up its game  in creating consistent characters. First off,   we need to make a description for a  character we want to be consistent.  Same with Leonardo, let’s use the prompt  “A realistic photo James, a realistic green   lizard-human guy, 18 years old. But most of  his features are human-like. His eyes are   curious and brown, and he's got a cool jawline.  When he smiles, it's friendly, and his nose   is uniquely shaped. Plus, he wears a stylish  jacket, making him a chill humanoid lizard.”  The results look promising. If you find an image  you like, you can click V1, V2, and so on to   generate more similar photos; let's save them. Now, upload the saved photos. After uploading,   save each image address by right-clicking  and save it in your notepad or chat box.  To make this character consistent, type “/prefer  option set” and set the option part as its name,   James. Click “plus one more,” and in the ‘value’  part, paste the image address we saved earlier.   Now, let’s try recreating the same prompt  we used earlier “Create an image of James,   performing at his own concert. Picture  him in a cool t-shirt, surrounded by fans,   with vibrant stage lights.” add double hyphens  then James, then double hyphens and v 6.  By providing Midjourney with reference images, it  should be able to reproduce James' facial features   in various prompts. Let's see how it goes! Okay, first of all. The face looks really   promising, and the images are spot-on with  those extra touches. Let's give it another   shot with the same prompt we used before, and  once again, it doesn't let us down. The face,   the background, and all those extra details  – everything's just perfect. When it comes   to making consistent characters, Midjourney  consistent character v6 is the go-to, no doubt.  Now, let's wrap up with a bit of a showdown. We'll  try making video snippets using these tools. For   Leonardo AI Alchemy, its “Upscaler” feature can  take a still image, pump up its resolution and   details, and then you can turn it into a short,  looping video clip with subtle movements and   animations using “image2motion.” Perfect for  adding a bit of life to your artwork or photos.  I tested it on the picture of James rocking  out at the concert, and after about a minute,   we've got our clip. It looks good, really  great. But there's a bit of an issue – some   facial features are a bit off throughout the clip. Now, onto Midjourney v6. The process is pretty   simple. Just add double hyphens and "video"  to every prompt where you want a quick clip.   But don't forget to finish with double  hyphens and “v6” at the end. For instance,   let's try making “lizard slowly catching a bug  with its tongue timelapse” – add double hyphens,   “video,” “v6,” and hit generate. Once your grid is  good to go, give it a thumbs up with the envelope   emoji, and the Midjourney Bot will send you a  link to your awesome video in Direct Messages.  But here's some cool news! Midjourney v6 is  gearing up to train video models starting in   January 2024, and they're planning to tackle  3D training after that. It's something to look   forward to! But in this video, it's clear that  Midjourney v6 is elevating its performance,   especially in handling complex prompts. While  both platforms deliver high-quality images,   Midjourney V6 takes it to the next level.  On the flip side, Leonardo Alchemy stands   out for its user-friendly interface. If you're a  beginner, Leonardo AI is definitely worth a try.  Once these features are ready, we'll make another  video to show you what they're all about. So,   if you enjoyed this video and found it  helpful, be sure to give it a thumbs up,   subscribe, and drop a comment below.  Stay tuned and watch our other videos   for more straightforward reviews on  the latest happenings in the AiWorld.
Info
Channel: AI World
Views: 746
Rating: undefined out of 5
Keywords: image to image leonardo ai, live canvas leonardo ai, leonardo ai live canvas, leonardo ai canvas editor, leonardo ai canvas, midjourney consistent character, how to use leonardo ai, leonardo ai realtime canvas, how to add midjourney to discord server, consistent character midjourney, midjourney product photography, midjourney seed, how to use leonardo ai realtime canvas, midjourney v6, midjourney 6, midjourney style tuner, leonardo ai, leonardo ai image to image, midjourney
Id: HhAVeQIqmbY
Channel Id: undefined
Length: 9min 3sec (543 seconds)
Published: Wed Jan 24 2024
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.