Gone are the days when you needed advanced editing experience and a big budget to create professional-quality videos. A new wave of AI-powered video generators has arrived, enabling anyone with a laptop and an internet connection to create engaging and captivating videos with just a few clicks.
Leading the way is OpenAI's Sora, an AI that can instantly generate several minutes of photorealistic video. That will happen when it actually hits the market. Until then, you can try out these innovative AI tools for free to easily bring your cinematic ideas to life. At the very least, it's a lot of fun to try out.
Hyper 1.5
Haiper 1.5 was developed by a team of ex-Google DeepMind researchers and is designed to compete with the likes of Sora and Runway's Gen-3 Alpha model. The AI generates two- or four-second-long clips based on input text, images, or other videos, taking about two to three minutes to generate once it reaches the head of the generation queue.
The system is easy to use: just choose a template (text, image, or video to video), fill in the prompts, upload your image or video, click Create, and wait for the AI to get to work. You'll need to do some editing and iterations to make the clip look and sound good, though, which will require you to hone your prompt-engineering skills. Because I asked them to create “a dog running across a green field toward a castle in the distance, in the style of a Ghibli movie.” Spirited Away” and this is what I got back (you can see the clip here):
It's currently free to use in beta, but like most generative AI platforms, the free tier is quite limiting, making multiple iterations of an idea a big challenge. In this case, free users can create 10 videos per day, three videos simultaneously, and receive 300 “credits.” Upgrading to the Explorer tier for $8 per month (billed annually) gives you unlimited daily creations, five simultaneous creations, 1,500 never-expiring monthly credits, and the ability to download videos without watermarks and set sharing status to private on Haiper's platform. The Pro package, which costs $24 per month, offers all of this, plus unlimited creations, 10 simultaneous creations, and 5,000 never-expiring monthly credits.
Pika 1.0
Pika is a free-to-use AI generative tool from the startup of the same name. It offers a suite of tools centered around a generative model that can create and edit videos in a variety of styles, from animation to cinema. The free AI video generator can create clips up to 3 seconds long (24 fps) based on the user's text and image inputs. Users can also edit existing videos to change the overall style, add effects, adjust the content within the frame, and adjust parameters such as the aspect ratio.
Pika isn't difficult to use, but creating clips longer than a basic three seconds requires some learning, as does Haiper. You can type prompts into a window or let Pika auto-suggest styles and themes to try out for you. You can also add sound effects manually or let the platform insert them automatically. The generator can also lip-sync your audio to an on-screen avatar, expand the canvas, or focus on specific areas of the generated video in subsequent iterations. If three seconds isn't enough, you can spend credits to add four seconds to your video at a time.
I gave them the standard image of dogs running across a green field towards a castle in the distance, in the style of a Ghibli film. Spirited Away” and waited two minutes. The system returned four potential video clips instead of one to review and further refine. Overall quality was roughly on par with Haiper, with more visual hallucinations and artifacts, such as a dog's tail disappearing and reappearing, and too many legs.
With Pika's free plan You will get 250 initial credits to use for generating videos, adding more time to videos, etc.Once the initial stock is depleted, Reddit will replenish up to 30 per day, Lip sync audio costs 2 credits per credit. Standard plan for $8/month includes: Features include 700 credits renewed monthly (with the option to purchase more at a discount), free lip sync, video download, resolution upscaling, watermark removal, and video length extension.
For $28 per month you get unlimited service and special offers.“The Infinite Chill Generation” You get 2,000 credits that renew monthly (30 credits per day if used up) and the perks of the Standard level. The Pro level, which costs $58 per month, pushes you to the front of the queue in Generation (also known as “Lightning Generation”) and lets you play with unlimited credits.
Canva
Canva is a powerful, easy-to-use digital design platform that lets you create everything from business presentations and brand logos to social media graphics, videos, and GIFs. Users can create projects based on the platform's templates and style suggestions, or start with a blank canvas and add text, graphics, brand logos, uploaded images, or freehand drawn elements to their creations. Or, they can enable the Magic Media feature, powered by Runway, and let Canva AI generate images and video clips ranging in length from 30 minutes to 2 hours.
Each generation request costs 1 credit, and free users can only get 5 credits. The system is very strict about the artistic style you can use, so you can't create anything like “a dog running across a green field towards a castle in the distance, in the style of a Ghibli movie.” Spirited Away” didn't actually work, but the system was Spirited AwayWithout these terms, the AI was able to return the horrifying results shown above in just a minute or two. It certainly needs some fine-tuning, but unlike the other apps mentioned in this guide, Canva lets you download your video directly as an MP4.
Canva's free version gives you access to the editor, 1 million+ templates, 3 million+ stock photos, thousands of design types from social posts to resumes, and 5 GB of cloud storage. A $20/month Canva Pro subscription gives you 100 million+ stock images, graphics, and audio clips, 20+ AI tools, and 1 TB of storage. Canva Teams, at $100/month per member (minimum of 3), helps businesses manage their branding and marketing efforts all in one place with real-time collaboration and design approvals, AI-generated branding copy, and AI management controls.
Base Lab
Based Labs is an image and video platform with cutting-edge AI tools that can generate everything from text input to photorealistic images and videos to digital selfies and tattoo designs. Users can also leverage the platform's AI-enabled face and gender swap apps and tools to upscale and enlarge images for free. Based Labs provides users with access to nearly a dozen AI image generation models, including Stable Diffusion 3.
Generating images in Based Labs is easy: just sign in, click “Generate,” select your model engine, number of output images, aspect ratio, and fill in the prompts. The generation process takes a minute or two. Generating videos in Based Labs is much more complicated, mainly because the platform leaves it largely to you to figure out the process yourself.
First you need to generate an image using the steps above. Then you need to bring that image back into the system (or just upload an image you already have), open it in the editor, and select Video from Image in the drop-down menu above the top-left corner of the image on your screen. It took me nearly 45 minutes of fiddling, Googling, and swearing to figure out the right technique, but I think that's a small price to pay for being able to generate (and download) professional-quality images and videos for free.
Entering the test prompt “A dog running across a green field towards a distant castle, in the style of a Ghibli movie” Spirited Away” I got the above image back, which I fed back into the system to create a video, resulting in a dreamy four-second clip of dogs running, swaying, and melding together. The system got the basics of the prompt right, but there's very little Ghibli influence to be found.
Each production costs 1-5 credits, with free users starting at 13 credits. A creator subscription costs $25/month. 1,000 credits per month, and the Pro package, which costs $49 per month, gives users 3,600 credits per month.
InVideo AI
Invideo AI is an AI video generation tool from ElevenLabs that is as simple and easy to use as Based Labs is complex. Users simply log in and click[AI ビデオの作成]choose either the Invideo 2.0 or 1.0 model, and enter the text prompt (up to 25,000 characters). However, the creation process is by far the slowest of the tools tested in this guide, taking over 10 minutes of actual load time before the screen prompting you to select your audience, appearance, and platform appears, and then another 3-4 minutes to actually create your video.
That said, the resulting videos are over a minute long and can be viewed in 480, 720, 1080, and 4K resolutions. Despite being less Ghibli-inspired than Based Labs, the AI has generously included both narration and music in the clips it creates.
Free users start with 10 minutes per week creating video content and can export four videos per week with an Invideo watermark. The Plus subscription, which costs $20 per month, offers 50 minutes of creation time per month (and unlimited exports), access to 80 iStock images per month, 100GB of cloud storage, and two audio clones. The Max package, which costs $48 per month, offers 200 minutes of creation time per month, 320 iStock images per month, 400GB of storage, and five audio clones.
Runway
Runway is an AI video generator based on the Gen model developed by the company's founders Cristóbal Valenzuela, Alejandro Matamala, and Anastasis Germanidis. It offers a range of AI tools, including text and image-to-video conversion, video-to-video editing, generative audio, and text-to-image tools. Users can generate clips of 5 or 10 seconds in length and 720p resolution. Prompt engineering is also a key skill here: the system can incorporate various storytelling aspects, such as the desired subject, scene, lighting, and camera movements and transitions. As with most generative AI, the more details and instructions you include in your prompts, the more accurate the resulting content will be.
Like Based Labs, Runway requires the user to first generate or upload an image and then create a video from that image. I used Runway's text-to-image generator to create the image above, then sent it back to the image-to-video generator with additional instructions for how the characters and camera should move. The result was more face-meltingly psychedelic than Ghibli-esque.
Unfortunately, the free plan only limits users to the older Gen-2 model and limits them to 500 credits per month. However, the $12 standard subscription gives you 625 credits per month (enough to create 10 seconds of Gen-3 Alpha video, 15 seconds of Gen-1 video, and 16 seconds of Gen-2 video), plus the ability to upscale videos between Gen-1 and Gen-2 models and remove watermarks from content generated by any of the three models. The $28/month Pro plan gives you 2,250 credits per month, and the $76/month Unlimited plan offers, as you'd expect, unlimited monthly credits, plus a host of other perks.
Final thoughts
Each of these six AI video generators offers users a slightly different experience and features. Choosing the one that's right for you will largely depend on what you're going to use it for. For example, Invideo is specifically designed for generating walkthrough and explainer videos for YouTube (and it even has a handy workflow plugin for that), while Haiper and Pika are better suited to creating short-form content for social media posts.
Canva doesn't produce as high quality (and low hallucination) images as Based Labs, but it's much easier to navigate and use, so Canva is better for hobbyists and casual users, while Canva is better for professional designers. The best part is that all these AIs are free to use. Try them all out and find the one that works best for you.
