Want to create those fancy Generative AI videos? Here’s a list of the best services that let you bring your imagination to life
Representation Pic
Open AI has blown everyone’s mind with its generative video model, Sora. However, it is not likely that the model will be available to try for free. Besides, it’s not available for the public yet, even if you are paying for premium access. But that’s okay because generative video will be big this year, and OpenAI is not the only player in town. Here are some options if you want to generate some videos of your own for free.
ADVERTISEMENT
Runway ML: Runway ML is possibly the best competitor to Sora out there right now. You get 125 free credits to play around with before you have to pay for pro access. You can use videos or images to get runway started. It might not be as good as Sora at text-to-video, but it’s just a matter of time before they get there. The 125 credits are minimal, so plan what you want to make before you do it. The lowest tier only gets you 600+ credits instead of the 125 offered, which seems a bit expensive. For your money, though, Runway ML delivers a whole plethora of options to create videos with a lot of prebuilt models. It is by far the best on this list. If you want to see what Runway is capable of, check out their Studio section, it is very impressive.
Pixverse: Pixverse can create text to video; however, the service works better with picture-to-video conversions rather than text. It is free for now, and there doesn’t seem to be a paid tier. That said, you will need to register with your email address. The output of Pixverse is a video for a few seconds. It lets you choose the style starting with the image and even a negative prompt to get the best you can. The opening webpage of Pixverse has a lot of examples of things that have been made on its site, and they are very impressive. I did try making some videos, but they weren’t as impressive as their homepage suggests.
Pika: Pika labs is the next impressive platform on this list after Runway. It can produce impressive three-second clips from still images, videos and prompts. The resulting video is impressive, to say the least. There are artefacts here and there, but overall, it has a good track record of producing coherent videos. You can also pan zoom and do all sorts of camera motions to get a realistic output. The free account is pretty good for Pika, and for the more serious connoisseurs, they do have a paid version.
Stable diffusion video: If you are looking for something that is absolutely free, then Stable Diffusion Video is it. You will need a robust machine with an impressive graphic card to run it, but once configured, it can produce very impressive videos. With Stable diffusion video, you control everything. This includes the style, the prompt, the input and the output. The unfortunate thing is that setting up a machine to run Stable Diffusion video is a little more complicated than it should be. There are, however, a lot of tutorials on YouTube showing you how to set up and use Stable Diffusion Video in a step-by-step way.
Moonvalley AI: Moonvalley goes the Discord route to produce its impressive videos. You basically have to join their Discord server and then enter the prompt there much like Midjourney. The videos take a while to generate though, much longer than Midjourney takes to generate a picture. I found the interpretation of the prompts not consistent, but looking at the prompts of other people on the Chat, it seems like you need to elaborate a lot for just a few seconds of video. The documentation of what works and what doesn’t is sparse, so mostly you are on your own figuring out how everything works. It isn’t as plug-and-play as some of the more direct web options here.
LensGo Ai: LensGo technically shouldn’t be on this list, because it does one thing well when it comes to video. In every other aspect, it is an image generator. Lens Go can take any video you have and transfer any style you want to it. Once you register, the site gives you 100 credits a day, and you can generate up to five videos a day with it. That said, because it is so one dimensional when it comes to making generative videos, it is possible that you may get bored with it fast.
![](https://www.mid-day.com/assets/images/flight.png)