Microsoft’s FREE AI Art Generator VS Midjourney V5

AI image generation is getting pretty Advanced and Bing just launched their own image generator to complement their new chat offering that they've just launched a couple weeks ago is this thing any good and how does it compare to mid-journey and the existing Dali well let's check it out you can go to bing.com create and I'm using Edge by the way just in case this is brand new it's in preview I might not even have access to it let's try it out a Macbook sitting on a desk in a brightly lit room join and create okay it says please wait your images are currently in progress I guess this is working now if you come to this page initially you might see like a waitlist type of situation which is pretty common these days for these AI tools and that's what I did initially but now that I came back it knows who I am it's creating an image for me apparently and it's got this 10 up here create images more quickly with Boost if you run out image generation may take longer I have 10 boosts it looks like a lot of people might be hopping on This Not That the news is out of the bag and it says thanks for your patience your images are on the way but it's taking longer than expected now I don't have a time frame when this is going to be fully released so keep an eye out on that but what I really want to see is iterative approaches to generating images with text what I mean by that is in chat GPT for example you can create a prompt it'll generate some text and then you have a sort of context that you're working with that the chat is aware of you're aware of it and you can create modifications to the previously generated text you can keep going further and generate more text based on what's already in the context I want to see that come to images because at this point in time there isn't a tool that does that and I'm hoping that Bing chat or being image Creator will change that where you can create an image from a text prompt just like you can with mid-journey and with other ones like Dolly and then you can say well I like that but can you modify it just a little bit can you move this little part of the image here can you just recolor this part of the image now Adobe just announced their Firefly project which I'm also on the wait list for all these wait lists but that looks pretty promising I can't wait to try that out by the way do you use any of these tools let me know in the comments down below I myself use mid-journey most often uh because so far it's had the best results dolly was pretty impressive a year ago but it hasn't moved on the news is that this image Generation by Bing is using the new version of Dolly underneath the hood I don't know what version that is Dolly 2.5 Dolly three um I don't know but it's not a new official version of Dali that you can use by going to Labs so here's labs and the images that it generates were pretty impressive a year ago but this stuff moves really fast so they're not so anymore and I'm afraid Dali hasn't really kept up not like mid-journey has I had to open a new tab because that one just uh froze for a while here it is on a new tab Same prompt being generated and now I have nine boosts well the progress bar just came to an end but then it jumped back to the beginning this is new stuff folks okay it's exciting but it also might have a little bit of a growing pain going on here we're getting used to that aren't we well darn I am going to show you something you watched this video for a reason and you want to see something and if this thing is not gonna work I'm gonna try the surprise me button uh my create button is no longer a button and it's still a button it's just deactivated and my surprise me button doesn't work either okay it finally finished and uh yeah that took a really really long time and this one the first one is still not done the second one finished and uh the third one just never gave me an option to create anything because I guess it detected that I was already running two queries so it didn't do it here is the result the Apple logo looks kind of chopped up it's using some kind of really interesting lens most of the image is blurry it puts a little Bing logo at the bottom there easy to get rid of but still annoying it is a brightly lit room I wouldn't call these great images though just to give you an idea of what mid-journey creates here are some MacBook pictures I've been creating one of these was actually in my thumbnail recently I'm gonna use the same exact prompt and run it through mid Journey version 5. and I'm also going to use Dali I do have some credits left on Dali so let's uh let's use them because I probably won't be using them too long and we might as well see who finishes first now mid Journey version 5 is quite a bit slower than version four you used to have to generate the images it would create smaller versions and then you would upscale it but now it's doing all of them in large scale right away and that's what's taking a little bit longer that's how I would explain that away mid-journey is done and Dali is also done if I look at the dolly images they're actually very very similar to what's generated by Bing create and the Apple logo is definitely even more messed up here I would call these pretty unusable let's take a look at the mid Journey version the desk the room these are a lot more clear a lot more realistic the MacBooks themselves well this is actually one of the older models of MacBook right here here's the 16 inch MacBook Pro they're all older varieties of MacBooks but they look more like MacBooks I'm impressed that the keyboards is not messed up the keys are actually they look pretty decent they're not per perfect but they look much better than Dolly's version it never gets the words MacBook Pro or MacBook correct but the scenery is generally way better in my journey I'm hopeful for this technology to improve but so far from what I've seen Bing image Creator is not that impressive unless you're doing maybe more artistic stuff here's some examples on their homepage surprise me should generate a prompt up here for you automatically it populates the text box and then you can click create and then wait for it to finish yet again did a pretty good job here with this surprise me the number two made out of leaves on a teal background photorealistic it's not realistic it's just photo realistic let's see how mid Journey does here we go mid Journey on the left and Bing on the right which one do you think is better maybe Bing is really good at this kind of things I I think actually brings us a little bit better majority thought that the leaves should also be teal apparently I provided only one color in the description I could go a little further and refine it but we're talking about a teal background not teal leaves so in this case Bing did better I would say for photos go to a mid-journey for now version 5 for things like this abstract art photorealistic art maybe Bing is good you want to see more of these let me know in the comments down below give this video a thumbs up and I'll see you in the next one

Channel: Alex Ziskind

Views: 43,228

Keywords: bing, bing chat, ai, chatgpt, gpt-3, gpt-3.5, gpt-4, gpt4, gpt, bing ai, openai, bing images, bing create, midjourney, image generation, image ai, generative ai

Id: 8IcgztZoNpc

Length: 7min 1sec (421 seconds)

Published: Wed Mar 22 2023