2026 OvenMediaEngine Roadmap

[getroot]

2026 OvenMediaEngine Roadmap

Happy New Year, everyone!
Thanks to your continued support, OvenMediaEngine has grown into what it is today.

First, we have some important news to share.

A New Chapter: OvenMedia Labs

Until now, AirenSoft had to run other projects alongside OvenMediaEngine to sustain operations. Even so, OvenMediaEngine has always been our core product, and we have consistently listened to and reflected feedback from the community. That said, it was unavoidable that our resources were divided.

Now, we have made a clear decision.

In Q1 2026, we plan to change our company name from AirenSoft to OvenMedia Labs and dedicate our efforts more fully to the development of OvenMediaEngine.

We ask for your continued interest and support as we take this next step.

---

About the 2026 Roadmap

This is not everything we plan to do in 2026.

We have a large backlog of ongoing improvements across stability, performance, and usability, and these are being worked on continuously. This roadmap highlights only the major milestones we want to share with the community.

Stability Improvements – Toward 1.0

Streaming servers are often used in mission-critical environments, where stability is non-negotiable.
So far, we have relied heavily on manual testing. In 2026, we plan to build internal tools for unit testing and end-to-end (E2E) testing to automate as much of our regression testing as possible.

Manual testing will still be necessary for complex and edge-case scenarios. However, increased automation will help us catch issues earlier and spend more time on the cases that matter most.

Note: The automated testing tools we plan to build are not just test code; they may need to work in close combination with properly configured infrastructure and third-party tools. Because of this, they may not be open-source projects. Even if these tools remain internal, their results will be continuously reflected in every OvenMediaEngine release.

With this improved foundation, our goal is to release OvenMediaEngine 1.0 in 2026.

Performance Optimization

Performance optimization is an ongoing effort for OvenMediaEngine.

In 2026, we will continue to improve performance across the entire pipeline, including media processing, protocol handling, and resource utilization. Many of these improvements are incremental and task-level, but together they play a critical role in achieving stable, efficient, and predictable behavior in production environments.

AV1 Support

AV1 was part of our plans in 2025, but at the time, the ecosystem did not yet feel ready.

Now, we believe AV1 adoption has reached a more mature stage. In 2026, we plan to add full AV1 support to OvenMediaEngine.

This is a large effort that includes both encoding and decoding across multiple protocols such as E-RTMP, WebRTC, and SRT.

Logging Overhaul

We know the logs can be overwhelming.

Over time, extensive debug logging has made it harder to quickly find the information that matters in production environments.

In 2026, we plan to significantly improve the logging system to make critical information easier to find, while still supporting detailed inspection when needed.

New Providers / Publishers

We are open to adding new Providers or Publishers, such as an SRT Pull Provider, if there is strong demand from the community.

If you have a concrete use case, please share it with us in the Discussions section.

---

Community Ideas

Now that we can focus more directly on OvenMediaEngine, we are able to spend more time reviewing and prioritizing community ideas.

Some ideas may overlap with Enterprise-only features. In such cases, due to contractual obligations or other practical constraints, we may not be able to provide those features as open source. We appreciate your understanding.

That said, our principle is clear: commercial products or integrations may remain proprietary, but improvements related to the core nature of streaming itself will be open-sourced whenever possible.

Our focus will be on ideas that bring real value to the widest range of users.

---

We always appreciate hearing from the community.
Thank you for continuing to support OvenMediaEngine.

[naanlizard]

Happy New Year!

The rebrand sounds good! I'll miss "Airensoft" though

The biggest feature I'd request is a way to have files saved to disk or ramdisk - especially screenshots and llhls parts. This would greatly simplify caching and generally serving files with e.g. nginx for us, I think.

Otherwise we're very excited for performance improvements and logging cleanup, sounds like this will be a good year!

[ysle]

Really like the direction of the roadmap. The strong focus on stability and performance further improves competitiveness. In that context, SRT Pull Provider support would genuinely simplify some typical workflows and could reduce overall pipeline complexity, which is also a strong selling point.

Happy New Year and wishing you continued success in 2026!

[nums]

OME is becoming a very complete solution, and it’s a great strategy to focus on stability.

For our use case, there is no need for major new features.
This is our list in priority order :

• Transport-CC bandwidth detection algorithm — I know it’s comming soon 🙂
• [o] Like @naanlizard suggested, saving chunks to disk to enable push LL-HLS or HLS would be a great improvement
• Subtitles for legacy HLS (I saw this in a recent commit 🚀)
• [o] Legacy HLS with multiple audio tracks
  => working with ?_HLS_legacy=YES with LL-HLS publisher
• WHEP — not mandatory on our side, but it would be nice, or at least to have a way to consume WebRTC streams in https://github.com/OvenMediaLabs/OvenLiveKit-Web
  (I tried to do this a long time ago in a PR)
• RTMP Listener/Publisher (nice to have): some of our clients still have workflows based on this. Replacing it with SRT could have some impact (latency, etc.), but it’s not mandatory at all

Thanks for all your great work !

[nums]

I tested the concept of live HLS and LL-HLS dumping, and here is my feedback:
I finally decided to build a separate script to handle live HLS push, as it turned out to be simpler for us to manage this way.

I also believe that LL-HLS cannot be easily dumped for live streaming, mainly because it relies on URL parameters. In contrast, standard HLS is quite straightforward to implement with just a few lines of code.

For that reason, I don’t think it is particularly valuable to integrate this directly into the OME core.

I may also share this feedback in the roadmap discussion to provide some insights based on my experience.

@naanlizard, If there is interest, I can ask my company to open source this component.

[ericsouza]

Ome is already such an incredible piece of software. Thanks for sharing the roadmap with us!

For our use case, the main missing features are WHEP and Recording Hooks.

I wrote "hooks" instead of "webhooks" because being able to execute a binary or script would be amazing. For our needs, we have to upload the video to a bucket and send a message to a broker. But, with webhooks, we could run a small sidecar HTTP API to handle that job.

Also, having a way to save the absolute timestamps to the recorded video (I am using ts) would be amazing, this way I could sync two different livestream from different cameras at same place.