AI Video Generators: Tech Review
Over one billion. That’s how many people are actively using TikTok every month. Out of every single person on the planet, one of eight people is on TikTok. Out of all internet users, a little under a quarter of them are on TikTok. In the United States, nearly half of the population is on the app, and the adults spend nearly an hour every day watching the trendy short-form videos.
These statistics from Backlinko.com show the sheer power of advertising that just TikTok holds. This does not include other video-forward platforms including YouTube Shorts, Instagram Reels, and most algorithmic feeds. Targeted feeds like can be found on TikTok also give an unprecedented boost to the advertising opportunities given to small businesses. The For You Page and Discover Feed allow small business and other entities with little platforms to compete with larger companies and creators that have massive promotion budgets.
When creating content, TikTok users can spend anywhere between 30 minutes to ten hours on a single video. While some users post “candid” content that is seemingly created in the moment and requires little to no editing, others spend hours painstakingly researching, scripting, filming, recording, editing, and promoting their video. The question on everyone’s mind, however, is how these trends can translate into being feasible for a business. Most businesses can’t get away with posting only candid and unedited content, but it’s unrealistic to expect anyone to spend ten hours on a single video regardless of how good the potential reach may be.
That’s where AI comes in! With AI video generators, the process of creating a viral video can be placed into the hands of a computer, the end result needing minimal human editing before posting. This way, the time saved can be focused on promoting the created video or brainstorming ideas for the next one.
Unfortunately, the market is flooded with different tools designed to incorporate AI into video content creation, and not all of these tools are created equal. A Google search for Generative AI tools yields too many results, most of which are lists written by the tools’ creators that unfairly bump that company’s tool to the top.
However, this list is created completely impartially. I tried over a dozen different tools from the perspective of a marketing team or small business owner looking to use the tool for content creation and digital marketing.
I judged the AI video generators based on these criteria:
At the bottom of the page, you’ll find a chart outlining all of the different points of comparison.
Best In Class: Veed.io
Veed.IO was one of the last AI video generators I reviewed, because it was relatively hard to find compared to the other products on this list. This is a shame because I loved Veed and could have saved myself a lot of time had I found it sooner rather than later.
Before I get into the official criteria, I would like to note that one of the biggest reasons that Veed earned Best In Class is simply how easy it is to use. The browser-based system has an interface that feels like a standard video editor and includes several similar features to a standard video editor.
Anyone familiar with software such as Adobe Premiere will feel at home in the Veed interface. That being said, however, it is also remarkably intuitive. Nearly every button includes a description on its function. Those buttons that are missing descriptions have helpful tooltips that appear when you hover the cursor.
Veed offers three different plans: Basic, Pro, Business. These run at $18, $30, and $59 a month, respectively, when billed annually. These prices are per user, although it is recommended that teams of any size start with the Business plan as the Basic and Pro plans are designed for single users. There is also an Enterprise plan with custom pricing based on business needs.
Determining the render time for Veed was challenging. It is very obviously faster than most of the other AI video generators, but the way the system renders clips makes it difficult to tell how long a clip is taking to render in relation to its size.
Unlike other AI video generators, Veed individually renders the avatars, allowing the rest of the video to be edited while the most recently added scene renders. Additionally, a 4 second clip only took about 30 seconds to render while the system was being used for other things.
The results are what one would expect from an AI video generator. The images and video provided by the software are crisp and the voices are smooth. There are definitely better voice generation softwares out there, but when used in the context of a full video, the quality isn’t noticeably poor nor is it noticeably great. The quality of results produced by Veed directly depends on the user’s skill and the extent of AI integration.
For users with a ChatGPT 4 plan, Veed offers a Custom GPT with integration directly into the OpenAI interface. This creates videos using the natural language prompting that we’re used to and allows users to edit it in the Veed interface.
For users on the Enterprise plan, brand assets such as fonts, music, color, and logos can be stored directly in Veed, making them easy to access for each of the content creators on your team. Consistency is key and the Brand Kit option offered by Veed ensures consistency in your team’s content
Veed offers hundreds of pre-made Templates for your videos, as well as the option to create your own templates. These templates come in a variety of styles and include different features, such a AI avatars and platform-specific sizing. The custom templates are great for users who are building a brand image and seek consistency across videos.
Arguably my favorite feature for teams, the Spaces offered on the Enterprise plan allows a user to have access to all of the videos and assets for a specific project and allows those assets to be stored separately from other projects. This feature is perfect for companies that run multiple advertising campaigns at a time.
Veed Allows for easy resizing of videos according to either user specifications or the size requirements of specific social media pages. The resize screen also shows what the video will look like on a platform in order to avoid text being obscured by buttons. This ensures optimal display and engagement across various platforms.
Best for Unique Content Creation: Runway
As you read through the review of Runway, you’ll slowly see what I mean by the following statement: Runway is best for when you have a concept but no set script. This AI video generator is incredibly adept at creation, producing results that can take your breath away, but I found myself fighting with the AI video generator to try and force it to output something that aligned with my very specific vision. Runway is great for when you want to create something that no one else has and nothing else could create.
There are four available plans for Runway: Free, Standard, Pro, and Unlimited. The monthly cost of these plans when billed yearly are $0, $14, $28, and $76 respectively. The plans each include a set number of credits per month. The credits reset every month on your subscription day, but if you run low before the credit reset there is always the option to buy more. Additionally, the Unlimited plan offers the ability to use “Explore” mode to generate using no credits at all, only a longer wait time.
The render time on Runway was by far the most disappointing part of this software. It took a 4s video about 1.5 minutes to fully render. Combined with the lack of access to the Preview function, the long render means that it could take over ten minutes to generate one standard 30 second TikTok. This time assumes that the TikTik generated properly on the first attempt and also does not include time to film the base footage, editing time, or other time sinks. In some cases, it may be faster to just create the TikTok by hand.
Beyond the render time itself, Runway also operates on a queue system. Your video is put on a waitlist to begin generation. The time for this time in between joining the queue and generation beginning is variable. The higher your plan, the shorter your wait.
This is where Runway really shines, but there is a caveat. The results of both the video and voice generation are stunning! The audio generation is one of the most in-depth on the market (at least that can be found in combination video/audio generation software), and the results are smooth and easily mimic human speech. My favorite part is how easy it is to generate the same sentence spoken by multiple people and easily play each to determine which to download.
The caveat I mentioned earlier is concerned with the AI video generation: The videos don’t always match the prompt super well. The prompt for the video above was “A girl walking on a side walk in a residential neighborhood, in the style of hyper realism, she is talking to the camera, she is walking at a slow pace, using her hands to speak.” While I’m sure I could have redone the video and prompt a number of times to get a better result, Runway is a credit-based system and even the Pro plans have a limited number of included credits. The render time is also quite long to have to redo generation several times.
For the sake of a thorough review, I also asked the AI video generator for two different image to video creations. The results are polar opposites, but I fully admit that it could also have been a prompting issues. Prompt image credit goes to D&D Beyond and @nadeemhamidy
Runway offers the ability to create custom AI video generators and train them with your own content. This is used to create anything from branded portraits to custom characters with Text to images. These models can then be used to create videos.
Runway offers 3D AI video generations, both capture and texture. This is a feature that I was unable to find on any of the other AI video generator that I reviewed. The 3D capture is still in Beta testing, so the results may not be great but the company is always improving. The 3D texture creator, however, works beautifully and renders incredibly quickly.
The best part of Runway is that in addition to being a software company, there is also Runway Research. This means that their software and capabilities are always constantly improving at a much faster rate than other companies.
Best for Jumping In: invideo AI
I found invideo AI through their Custom GPT. Until I discovered Veed.IO, I was under the impression that this system was the easiest to use. That being said, invideo is the easiest for someone who has never used video editing software before.
The user interface is fairly barebones, which just means that it’s easier to use due to the lack of visual clutter and no fancy tools to get confused by.
There are three base plans for invideo, but there are different tiers within the base plans. The prices range from their free plan all the way up to $480 a month for the Max plan with 10 users. The base prices for the Plus and Max plans are $20 and $48 a month when billed annually. These base prices include 1 user each. As the price and number of users increase, so do the included minutes of AI generation per month, the included iStock pictures, and the included amount of storage.
invideo is one of the more expensive plans on this list. For a small business that is just starting out with video content creation, though, it has one of the most generous free plans on the list. The free plan includes 10 minutes of content creation a week, which is only ten minutes less per month than the cheapest Plus plan. The free plan does not include iStock, but you are able to upload your own media and it does come with standard media. The exported videos are going to have a watermark, but it is small and easily missed.
Where invideo truly shines is the render time. It took about 30s to render an 11s video, which already had the audio overlayed. It also offers an AI-powered editing option for after the video is created. This is a fairly unique quality for this type of AI video generator. The render speed also makes it feasible to just regenerate the video as many times as necessary until it meets what you’re looking for. Because it’s a time-based system instead of a credit system, regeneration is more feasible when using invideo than it is for other AI video generators.
The results from invideo are fine. The video certainly won’t take your breath away, but will be ready to publish with relatively little editing. This AI video generator also removes the watermark from stock images and gives you the license. The free plan generated this video, so it still has the watermarks.
The Custom GPT offered by invideo is where difficulty of use goes from ‘easy’ to ‘child’s play.’ The OpenAI technology found in ChatGPT combined with the video making powers of invideo means that it’s possible to create a ready-to-post video using natural language models.
Honorable Mentions: Elai.IO and Pictory
Where there’s first place, there must also be second place. Elai.io and Pictory both were amazing and impressive AI video generators. The tools they had that allowed them to shine just weren’t big enough to earn them a spot on top.
The unique tool that earned Elai a spot in the honorable mentions is definitely the ability to create a video based on an uploaded PowerPoint, slide deck, or PDF. This would be useful for marketing teams or small business that want to make content using resources that they already have made. The results aren’t amazing, but are far from bad. It does add the ability to feature someone talking and presenting the powerpoint, as well as adding animations to images.
Elai’s downsides include long render times, a clunky user interface, and fairly high prices when you consider the included features.
Pictory was one of my personal favorites when I first came across it. It earned a spot in the honorable mentions for not only the ability to create a based on an article or blog post, but the wickedly fast render times! It took Pictory 30 seconds to render 37 seconds of video! Veed was the only AI video generator that beats this speed and the two are still remarkably close.
Pictory’s downsides comes from the fact that the visuals all come from stock content and the licenses are not provided. The prices of the plans are also quite high. There is a free trial available for each of the plans, though
Dishonorable Mentions: Colossyan and Synthesia
And where there is first and second, there also must be last. I am including these AI video generators purely because they are some of the first listings on Google. Due to their apparent popularity, I feel the need to warn you against them.
Colossyan is often lauded as one of the best AI video creators out there. It may make more sense for individual content creators. For our purposes, though taking over 7 minutes to render 17 seconds of video doesn’t make sense. It is impractical at best and laughably inefficient at worst. Colossyan is expensive for the included features and takes forever to generate a fairly basic video. On top of this, the results don’t demonstrate the cost or time commitment involved.
Synthesia landed on this list by not offering a free trial. Different businesses and individuals are going to need different features and capabilities. As such, some softwares are going to work and some aren’t. Without a trial, you may pay hundreds of dollars just to find that an important feature is missing. There is a way to get Synthesia to send a “sample video” to your email. However, this sample is fairly limited and fails to demonstrate the user interface at all.
|Very fast, clip rendering
|4 second video renders in about 90 seconds
|11 second video renders in about 30 seconds
|6 second video renders in about 60 seconds
|37 second video renders in about 30 seconds
|17 second video took about 7 minutes to render
|Included (and voiceover uploads allowed)
|Included, but seperate from video generation
|Included, and easy to use
|Included, but only in the higher-priced plans
|Monthly Price Range (when Billed Annually)
|$18 – $59 or custom enterprise pricing
|$0 – $76 or custom enterprise pricing
|$0 – $480
|$23 – $100 or custom enterprise pricing
|$19 – $99
|$19 – $61
|$22 – $67 or custom enterprise pricing
|Rapid rendering; feels like a standard video editor; customGPT
|AI training; text/image to video generation
|customGPT, very basic and easy to use interface; licenses included
|Video from powerpoint or PDF
|Article to Video
|Cloned voice included with any yearly plan
|The results aren’t stunning, they’re just great
|audio generation and video generation are completely seperate, long render times.
|Visuals are almost purely stock content and not created by AI
|clunky interface; high price point for the included features
|Visuals are almost purely stock content and not created by AI
|excessively long render time, the results aren’t great
|No free trial
There is no one who can tell you what solutions are best for you and your company. It all depends on your needs, skill levels with AI and video editing, and budget. It’s also an option to use multiple AI video generators. and Runway, for example, compliment each other really well. Where each AI video generator is missing features, the other software offers it. That being said, combing through all of these AI video generators is still hard. You always have the option to trust the experts to incorporate AI video generators into your workflow.
At Online Optimism, we believe in using AI to innovate your business instead of replacing the human touch entirely. Interested in using AI to improve your content or your digital marketing game? Subscribe to our newsletter, read our blog, or check out our services page. We look forward to hearing from you!