I Tested Every AI Video Tool To Find Out The Best One

AI video generators have come so far over the past few months and partially that's due to the competition. There's really not one tool that just excels in everything at this point. And as Chinese models compete with Google, we the consumers of this technology keep getting increasingly more capable tools that open up new use cases that were not thinkable a few months back. And that's why I decided it's time to make an updated tier list video where we rank every single video generator that is worth knowing about along with some practical use cases for these. The way we'll be doing it is by placing them on a tier list. S tier being the best, A tier being second best, and so on. We're ranking by raw quality of the video generator and the features that the platform that you use it in as a consumer provides. We won't be taking cost into account here. We're all about use cases and quality in this comparison and we'll be doing it through a ton of examples and test cases. I have a thing or two to say here, but if you disagree with any of these opinions, I'm happy to discuss it in the comment section below.

And with that being said, let's start with the lower tiers in the list right here. Basically, me and the team could have put way more generators down here, but we sort of figured it's not even worth the time. There's a lot of different ones out there right now, but these two generators right here in Cindier are not really worth your

attention. So, what I'm going to do here is I'm going to skip ahead and show you the number one spot so we can compare the different ones that we'll talk about now to it. And if you have experience in the space, you might expect this. It's Google's V2. Now, yes, I think it could be argued if it's the best or the second best, but it's definitely up there in the S tier. And I placed it as first because I believe it's the most well-rounded one out of all of them. If you just had to pick one video generator, I think you're well served with V2. And now they're providing it in their various apps like Gemini Advance or Google AI Studio where, by the way, at the time of this recording, April 23rd, you can just try this thing out for free. That won't last for long, but you can also use it within other applications and it just looks fantastic. So, we'll be using it as the gold standard here, comparing it to some of the other options here. Now, I want

to move to some of the Btier models because these offer some unique capabilities that you might want to use in specific situations. And first up, we have Sora. So, this was obviously super hyped when it was first announced, but then a lot of the other companies, including some of the Chinese competition that we'll look at in SN8 tier, just overtook it in raw quality. I think the interface is very beginner friendly and the tooling is really solid. But the quality at this point just isn't even close to some of the top competitors here, and that's why it lands in Btier. Nevertheless, if you have a OpenAI subscription, you get access to Sora already. So, instead of having to buy a extra sub, you might just want to use that. And that's why we placed it in B tier because for people with that sub who don't want to spend extra money, it's a great choice. Next

up, we have Pika. And Pika kind of carved out this unique spot within the AI video landscape. It does these really funny effects. Some of them went really viral, like cakeifying any object. You can just upload your own videos and it will turn them into cake. There's many more of them, and their specialty is really adding these specific features that are super easy to use and just create these unique effects. Although the quality of the model might not be the highest, if you're looking for quick fun results or you just want to create some memes, Pika might be the go-to one, and that's why it's in B tier. Next up

we have Adobe Firefly. This one is Simply Explained. It's not the highest quality model, but it's the only one that says that 100% of the training data is copyright free. So, if you're really, really worried about AI generations being copyright safe, then this is the one you want to use. Although I should add many times at this point has it been clarified that no AI image or video generation is protectable by copyright. There have been some court cases where the precedent for this was already set. Nevertheless, if you really want to make sure that the videos you produce are safe to use, then Adobe Firefly is your friend. Next up, we have

Pix and the model is sort of okay. It's better than some of the C and D tier options or some of the other ones that didn't even make the list here. But also consider the sorting here to be a factor. So Pixver would be at the end of Btier. It can be surprisingly solid, but just not really great. That's really all I have to say about this one. Google's

V2 model obviously sits in S tier. But one of the most interesting ways out there to use this model is through the sponsor of today's video, which is LTX Studio. And it couldn't be a more appropriate sponsor because LTX Studio delivers a suite of tools that allows you to really utilize these models to tell a story or to create a sequence of shots with ease. And if you follow the channel at all, you know that we've been covering both V2 since the day of release and LTX Studio also since its release. And historically, I've always praised it for its tooling, but the one puzzle piece that was really missing there was a state-of-the-art video generator, which it now has with V2. So, let me just quickly show you what you can do with this powerful combo of a suite of tools of LTX Studio and Google's V2 combined. So, first up, there's the motion generator. And this is simply the easiest way to use V2. It uses an image as a starting point and by uploading that and entering a prompt to describe the motion that you want in a shot, you simply hit generate and within a minute you'll see the image come to life and that's really great. But what makes LTX Studio so impressive is how it integrates VO2 with the rest of the already very capable platform. And if you don't have an image, they have a built-in image generator so you can get started right within the same interface. Then you can take some of the images you created and use them in a storyboard. You can set up an entire project with custom images and then use V2 to add motion to every single one of them. This is really the workflow you want if you're going from just an idea to a nailed down visual concept in the minimal amount of time. So, if you have a vision and want to turn it into video, or you want to see what kind of short film or video you could come up with in collaboration with AI while using one of the state-of-the-art video generators out there, there's really no better way to get started than with LTX Studio. I do have to note that if you're going to be using V2, the Google model, you do have to pay per usage. And one amazing fact here is that if you're actually looking to generate with VO tool, doing it through LTX Studio is the cheapest way available today, but if you follow the link in the description, they're giving away free bonus credits to new users so you can test this brand new V2 integration that just launched. Thanks again to LTX for sponsoring this video. And now, let's move on to the next video generator on our tier list. All right

now let's start at the end of a tier. And here, after a lot of testing, we placed Mini Max, which was really impressive when it first came out, but at this point, I think some of these other models outperform it. But as soon as we get into this S and A tier territory, I want to reveal another use case that we've actually been using at the advantage all the time. You might have noticed, but we use quite a few AI generated clips as B-roll. Often, we generate an image of something we want to represent and then use one of these S or A tier video generators, like Miniax in some cases, to turn it into B-roll, which we then use to spice up our videos. This is a great and more custom replacement to stock footage and my personal favorite use case for AI video.

Another option you have obviously is runway. They really have been pioneering from the beginning and they have the most comprehensive set of tooling here because they've been doing it for so long. Their brand new model is now on par with some of the second generation models like Google Vo Minia and some of the other S tier models. But I just got to say from our testing we just found it's not quite there. But the tooling is really good and that's why it landed in a tier. And when I say tooling, I don't just mean image to video. That's sort of the standard requirement. I mean things like inpainting or lip-s syncing. Not all of them are available with the new model yet. But as per usual, they become available over time and then they upgrade to another new generation. Okay.

The next one is a really interesting and a super unique one on this list. It's the Alibaba 12. 1 model. And if you haven't heard about this, this entire model is open source. And by that I mean fully open source. So you can completely disconnect from the internet, run this locally, and start your own AI video production inhouse fully offline. because Alibaba just decided to put this out there and the quality really is on par with some of the best players here. Not quite as good as some of the S tier options that we're about to cover, but as you can see on screen right now, this thing is really impressive across many different disciplines. And along with the next one on the list, it's one of the two really great open source alternatives when it comes to AI video generation. The next one here is Hanuan

Video. Another Chinese open-source alternative that you have here. Very, very good quality. something that on release everybody was stunned by, but the space just keeps moving so fast. I do believe that some of the most recent S tier models that we're going to cover next do trump it in terms of quality. But then you can get really nitpicky here and it depends on how much you care about things like prompt adherence, which means if you give it a long prompt with a lot of details, is it really going to stick to exactly what you want from it? Well, that's where some of the next options really outshine it. Okay

now we get into the S tier, and these are really bestin-class ad specific things. I'll do my best to point out the specific differences here because you might want to use different models depending on what you're going for. But starting out with Dream Machine by Luma Labs here, particularly with the Ray 2 model that is now available through Dream Machine, this is really bestin-class when it comes to animations and motion graphics. Now, don't get me wrong, it's really great at other things too, but when it comes to really animating a logo or a title, something like that, Ray 2 nails it. The interface is really intuitive. You can get results like this which enhance any project that might have video in it. Heck, I would even go a little further. I think with the tools being this capable, you could and should be using this in any presentations you do. And this is sort of the next use case that I want to highlight here. If you're doing a PowerPoint presentation, why not spice it up with a little bit of animation that you create with AI video tools. That's something I've been doing more and more. Maybe opening your presentation with a little video slides that keeps looping, which is an option you have in Dream Machine here. You can just make the clip loop indefinitely while you make your point. Okay, on to

the next one. And honestly, this one could be in first spot. It really depends on what you're prompting for. It's the new Cling 2. 0 model. Now, it lacks a lot of the features that some of the competition here has, but that's only because it's just rolled out. And the model is ridiculously expensive, but again, we're not judging that here. The prompt adherence is unmatched. If you really have a long prompt and you want the AI video generator to nail it, Clink 2. 0 is your friend. The problem is that there's no API for this yet, and there's no unlimited plan. And if you ever worked with AI video, you might know that often it takes multiple generations to get exactly what you're looking for. And you having to pay multiple dollars for a single generation. This is just not a feasible option for most people. So the lack of flexibility really puts this in third spot here on this list. But again, if you only isolated the raw quality, this one could easily be in second or first place. And then the last

missing one here is Hicksfield AI, a video generator that recently released to the public. And this one really shines in generating humans, which also includes getting human anatomy right, which is something that a lot of these generators are not very good at. Good luck making shots of somebody walking or, god forbid, something more complicated like dancing. It just doesn't work reliably with any other tool here. Higfield actually nails this and it has a ton of camera presets where you can really precisely control the camera. Now, this camera control can be found in many other of these tools like Runway, but Hicksfield just does this best and most reliably. So, if you need to generate images of humans reliably, Hicksfield is probably your best choice, although the quality might not be best in some other categories like animation or creating B-roll clips. And then

lastly, as I pointed out previously, Google V2 for its overall quality availability through the API, userfriendly prompting, and just overall robustness takes the first spot. If you have to choose only one model on here, then you can't go wrong by just going with V2. By the way, I included links to all of these in the video description

below. And to round out this video, I want to show you one last big comparison graphic between all of these video models of a man with his dog, so you can make up your own mind on which one of these works the best. And as you do

that, there's one more use case I want to highlight, which is actually commercial usage of these. More and more advertisements have been incorporating AI video. And not just that, some of them decided to go full-on generative AI. And they did receive a lot of backlash in some cases, like this one on the screen right now that has been made by Coca-Cola back in 2024. received a lot of negative backlash because it did look a little suspicious to a lot of people. And yeah, fair enough. I think this era of fully AI generated pieces is not here yet. We'll get there eventually. But right now, we're really in this phase where integrating this into existing workflows, enhancing your presentations, creating funny clips, generating extremely customized B-roll clips or simply using it as an animation tool are all things you can be doing today. And all you need to do is pick the right tool, get one of these subscriptions, and then you're ready to go. just front some music, get generating and see how far you can push these things. All right, that's

everything I have for today. If you like this video, I want to highlight that every single month we update this tier list in the free space of our community. I will also link that below. So, if you ever wonder what's the best AI video generator right now, you can simply check out that free post that we update monthly, completely free. You don't even need an account. And if you want to stay up to date week to week, check out the AI news you can use show that I do every single Friday on this YouTube channel. All right, that's all I got for today. See you soon.