Revideo is an open source framework for programmatic video editing. It lets you create video templates in Typescript and render them via an API. It also provides a React player component to enable previews and real-time edits in the browser.
Thank you for the hunt, Michael!
Hi Product Hunt community! 👋
I'm Justus, and I'm building Revideo together with my co-founder @hkonsti. We're excited to show it to you today!
At the start of this year, we were exploring a bunch of different product ideas in the realm of AI-enabled video editing, ranging from A/B-testing and automating facebook ads to generating memes.
While building these products, we were consistently frustrated by the tooling available for programmatic video editing and therefore decided to build a framework ourselves!
Revideo lets you create video templates in Typescript and render them with dynamic inputs through an API. It also comes with a Player component that lets you preview your projects in the browser and integrate video editing functionality into web apps. Our renderer is built on top of the WebCodecs API, making video exports super fast. On my Macbook, I can render 60 seconds of full-HD video in just 14.3 seconds. For faster rendering speeds in production, you can also parallelize rendering across multiple AWS Lambda workers. ⚡
Revideo is MIT-licensed and will stay that way. In the long-term, we want to make money by offering a managed rendering platform where you can deploy your Revideo projects and render videos super fast.
If you want to learn more, feel free to check out our Github repository: https://github.com/redotvideo/re...
We are looking forward to your feedback! 😻
Report
@hkonsti@justusmattern Huge congratulations on the launch! Wishing you all the success in the world. What feedback have you received so far?
@hkonsti@kjosephabraham Thank you very much! Feedback so far has been pretty good - a few weeks ago, the most common complaint were slow rendering speeds, which are now fixed. As of now, I would say that we care the most about reliability across different environments and setups
This is so cool! Love to see new approaches to video editing. Also, noticed that the 'Basic code example' and 'Rendering Video' pages at the bottom don't work, congrats btw!
Congrats boys! This has been really fun to follow over the past months. Love the permissive license and put myself down on the waitlist for the platform.
Congrats on the launch @hkonsti and @justusmattern. I played around with it and it seems really cool (love that it is OSS)!
What kind of tooling do you recommend to generate videos rendered with revideo (e.g. cursor with indexed revideo docs)?
@marc_klingen Thank you Marc! Really appreciate it
Revideo docs are obviously not yet part of the training data of today's LLMs, so indexing the docs is a great idea! I personally like to use https://www.continue.dev/ for doing so, but have not yet tried it. Probably the most important tool if you're building Revideo apps is a hosting platform that lets you do parallelized rendering. We have used AWS Lambda and Google Cloud Functions for this, but I'd imagine that newer services like trigger.dev can also handle this. Soon, we'll also launch our platform, which will handle parallelized rendering out of the box :)
Report
Congrats on the launch, Justus! Do you have any detailed documentation or examples for integrating Revideo with existing React-based web apps?
@memphys_sk Thank you! Yes, our docs have a dedicated section on building web apps with Revideo, which explains how to embed previews and how to set up a rendering service as part of your web app: https://docs.re.video/category/b...
Congratulations on the launch! Idea: you could also include a voice-over text as a parameter. Something like . This would make the video that you posted above possible with Revideo itself :)
@semanser Thanks Andriy! The posted video was fully created with Revideo, you can check out its code here: https://github.com/redotvideo/la.... We have thought about integrating AI-specific components into revideo natively, and are still considering this, but have for now deferred it. It is relatively easy to run scripts for generating AI assets (such as text-to-speech audios) in a separate module of your app and then feed the response as a parameter in your revideo project.
AI-specific components would be somewhat restrictive and opinionated by design. For instance, what if you're building an AI short video app, and then your user wants to switch out a single asset that is AI generated? For these purposes, it's a lot easier to handle these things seperately in my opinion
Revideo
Revideo
Coinfella
Revideo
Revideo
Langfuse
Revideo
Langfuse
Revideo
Revideo
DepsHub
Revideo