AV2 Arriving: What We Know, and What We Don’t Know
According to AOMedia’s announcement, AV2 is set for release at the end of 2025. The group highlights several improvements over AV1, including significantly better compression performance (with no specifics), enhanced support for AR/VR content, split-screen and multi-view delivery, improved handling of screen content, and operation across a broader visual quality range. Like AV1, AV2 will be offered under AOMedia’s royalty-free patent policy, which means licensees can use it without paying royalties to AOMedia.
The release also included survey results from AOM members: 53% reported plans to adopt AV2 within 12 months of release, while 88% said they expect implementation within two years. AOMedia positions this as evidence that AV2 will follow AV1’s trajectory, which is already deployed at scale by companies like Google, Netflix, and Microsoft.
What We Know From Other Sources
We hunted for observations from developers who have been hands-on with the AV2 code and found none. As of September 2025, there are no published real-world encoding or decoding benchmarks, no playback performance measurements, and no field reports from device makers. What we did find are technical previews, developer commentary, and lab-stage analysis that sketch out what AV2 is expected to deliver.
Technical summaries note AV2 is expected to bring improved intra-prediction, enhanced temporal filtering, stronger handling of screen content, and support for multi-program split-screen delivery (VideoProc). These features line up with AOMedia’s claims but are described with more technical detail.
The draft specification is circulating and being refined, with AOMedia reiterating its plan to finalize and release AV2 by the end of 2025 (VideoCardz, Linuxiac). But the lack of open reference implementations or decoder builds means that developers have not been able to perform independent tests. As the Multimedia.cx developer journal put it in June, industry chatter is strong, but there are no “end-user relevant” results to report yet.
In short, the non-AOM sources confirm what AV2 is expected to be, but not what it is. The real numbers on compression efficiency, encoding speed, and playback requirements remain unpublished.
The IP Question
AOMedia stresses that AV2, like AV1, is developed under a royalty-free patent policy, with all contributing members agreeing to license essential patents without royalties (AOMedia press release). This means that you can license and use AV2 without paying royalties to AOMedia. But it doesn’t mean that AV2 will be totally royalty-free.
There are two pools claiming royalties on AV1 encoded content (Access Advance, Avanci) and one claiming royalties on consumer decoders (Sisvel). To the extent that AV2 borrows patented technologies from AV1, the same royalties may apply. There may be new technology also covered by third-party patents. The bottom line is that you should treat AV2 like you should be treating AV1; before implementing, check with your IP attorneys.
What’s Relevant for Deployment
The fundamentals for publishers are still unknown. There are no published figures on how much more efficient AV2 is than AV1, and nothing concrete on encoding speed or cost. Until those numbers arrive, it is impossible to judge whether AV2 will save bandwidth without blowing up compute budgets.
What we do know is that AV2 is a new codec, which means it will not play on existing AV1 decoders. It is also likely to be more complex than AV1, which means a heavier load for playback. The practical implications vary by platform:
Desktop
AV2 may play on recent computers once an optimized open-source decoder (the successor to dav1d) appears. Expect YouTube and Facebook to lead deployment in early 2026, since they control both content and playback environments. But premium publishers will hold off until DRM systems like Widevine, PlayReady, and FairPlay are fully integrated. That gap mirrors what we saw with AV1: UGC first, premium years later.
Mobile
If AV1 strained low-end Android devices, AV2 will be worse. Software decode may work on high-end phones, but performance and battery drain will limit adoption. Real deployment depends on hardware support from Qualcomm, MediaTek, and Apple. These vendors have invested heavily in VVC, which makes fast AV2 integration unlikely. Even Apple, an AOMedia founding member, waited until late 2023 to enable AV1 hardware decode, five years after AV1’s launch. A similar delay for AV2 would push efficient mobile playback into the early 2030s. As a proof point, AV1 finalized in 2018, and still isn’t widely available on mobile in 2025, more than 7 years later.
Smart TVs and Dongles
No current devices will support AV2, so new silicon is required. But the dynamics here are different from mobile. Netflix and YouTube exert enormous influence over smart TV adoption. Their demands for AV1 pushed Samsung, LG, and Sony to implement hardware decode faster than mobile SoC vendors did. If Netflix and YouTube push hard for AV2 in the late 2020s, smart TV and dongle support could arrive sooner than mobile, though meaningful market penetration is still unlikely before 2027-28. Again, AV1 finalized in 2018, and the first smart TVs with AV1 playback appeared in 2020.
Figure 1 puts these numbers to a schedule. As AI codecs become more real, dates beyond 2030 become increasingly speculative.
Figure 1. Estimates of when to expect AV2 deployments
Conclusion
AV2 is coming, and AOMedia is promising significant compression gains, broader quality range, and support for emerging use cases. But the timing of relevance will depend on real-world playback support.
If past is prologue, adoption will be uneven. As we sit here today, we are seven years post AV1 launch, with a pandemic and supply chain FUBAR in between. The pandemic might have accelerated adoption, but supply chain delays clearly slowed it.
In 2025, most new TVs support AV1 because no manufacturer wants to ship a TV that didn’t fully support YouTube or Netflix. AV1 hardware support in mobile may just be reaching critical mass, a huge chunk of that support comes from Apple and high-end Android devices, which handled AV1 without dedicated silicon. AV1 hardware support on $100 mobile devices is minimal due to margin issues. Desktop and mobile browser support from Google and Mozilla came very quickly, but DRM support was delayed. As a result, outside of the big three platforms, publisher support was limited.
The same dynamics are likely with AV2. Desktop browsers may see early software playback in 2026, led by YouTube and Facebook, but premium adoption will wait on DRM integration. Mobile is gated by hardware support, and with Qualcomm, MediaTek, and Apple invested in VVC, AV2 won’t reach critical mass there before 2030. Smart TVs and dongles could move faster than mobile if Netflix and YouTube push, but meaningful penetration is still years away.
As with AV1, the “royalty-free” label is limited to AOM members, and third-party patent pools may claim rights. Publishers should keep AV2 on the roadmap, test it when encoders and decoders appear, but not plan for large-scale deployment until device support is real. And before deploying at scale, consult legal counsel about potential IP exposure.
The big question to ask yourself is why AV2 will be different than AV1, particularly with VVC (and to a lesser extent, LCEVC) support solidifying and AI-based codecs near release. It could be different in new markets that AV2 is focusing on, including AR/VR content, split-screen and multi-view delivery. For general-purpose content, AV2’s value proposition looks uncomfortably close to AV1.
Related Articles
The arc of AV1 codec adoption at large-scale content owners has been long and complex, as Meta Senior Media Software Engineer David Ronca (who also spent 12 years developing encoding solutions at Netflix) knows as well as anyone. In this interview with Jan Ozer of Streaming Media and Streaming Learning Center, Ronca discusses the integration of AV1 into the Android ecosystem, focusing on the challenges and solutions for mobile devices.
21 Feb 2025
Scheduled to be the first codec released by the Alliance for Open Media, AV1 is positioned to replace VP9 and compete with HEVC. While we don't know many details yet, the backing of the Alliance should give AV1 a significant competitive advantage.
03 Jun 2016