The State of Streaming Codecs 2026
Streaming codec adoption used to be an engineering abstraction governed by RD curves, BD-Rate tables, and road map slides that no one outside of R&D ever considered. Over
the last 15 months, codec adoption decisions have morphed into a much broader discussion, involving C-level execs from finance and legal.
While the precursors to this transition occurred pre-2025, the situation coalesced in 2025. During the same period, we saw one codec step to the front (AV1) and another shrink before our eyes (VVC).
This year in review considers the following six events and discusses their implications for those involved in codec implementation decisions:
- Access Advance launches the Video Distribution Patent (VDP) Pool (Jan. 16, 2025).
- Nokia and Amazon settle their global streaming video patent litigation (March 31, 2025).
- Sisvel announces that it licensed roughly half of the AV1 finished-product hardware market (July 31, 2025).
- InterDigital acquires AI-native compression startup Deep Render (Oct. 30, 2025).
- Brazil regulates “TV 3.0”/DTV+ via Decree No. 12,595/2025 and adopts its next-gen TV stack (Aug. 27, 2025).
- The Alliance for Open Media (AOMedia) announces the AV2 next-generation video codec with a year-end 2025 launch (Sept. 15, 2025).
As you’ll see, while several of these events relate to codec technology itself, most concern who will pay for codec usage, and how much, in 2026 and beyond. Let’s jump in.
Access Advance Launches The VDP Pool
On Jan. 16, 2025, Access Advance announced its Video Distribution Patent (VDP) Pool covering content distribution using HEVC, VVC, AV1, and VP9, with rates published in July 2025. Royalties are calculated as the higher of three metrics shown in Table 1: average monthly active video users, average monthly video subscribers, or semi-annual streaming revenue. Significantly, Access Advance offers a royalty waiver that excludes many small streaming services from any liability.
Table 1. Access Advance VDP royalty rates
Note that only those subscribers who view a video encoded by one of the four listed codecs are included in the three metrics shown in Table 1. So, if a publisher initiated AV1 usage exclusively to serve Android users, rates would be calculated only for those users, not for the total subscriber base.
Access Advance wasn’t the first content pool; Avanci launched its Avanci Video Pool back in 2023. At press time, Avanci released partial rate information and announced that it signed its first licensee, which it didn’t name. You can read about the rate and licensee here. Significantly, at least for the rate options Avanci announced, based on revenue and a per-user charge, there is no cap at the top or de minimis exception at the bottom.
In contrast, the Access Advance pool has both an annual cap and (as mentioned) de minimis exceptions for smaller services. The pool has six licensees—five of which are also licensors—and Roku, which became a licensee in January 2026 and is not a licensor.
Confusing the royalty situation further is that the programs share 16 common licensors, which means that licensees of both programs get an offset for any double royalties paid. Both programs should know what that offset is, but it’s not listed on their respective websites. If you’re trying to estimate your potential royalty obligation to both pools, this offset is critical information to obtain.
The Implications of Content Royalties
Viewed broadly, new codec adoption is driven by two different factors; one is practical entry into a new market, like HEVC/HDR provided for 4K premium content. Streaming publishers in this category must use licensed codecs or drop out of the premium content race.
The other motivation is to harvest bandwidth savings from increased compression efficiencies. Once Access Advance’s royalty rates landed in July, the math for streamers deploying advanced codecs for additional efficiency changed immediately. In 2019, the AV1 or HEVC pitch was simple: spend 3x–5x more on encoding to save 20%–30% on CDN for all compatible endpoints. In 2025, for services above the VDP Pool thresholds, that same pitch becomes 3x–5x the encoding cost, plus a royalty that can reach mid-eight figures annually and (if you’re large enough) attract attention from individual patent owners like Nokia (see the next section). While the roughly 30% bandwidth reduction hasn’t changed, CDN fees are much cheaper, reducing the associated savings.
In short, larger services delivering HDR need AV1 or HEVC for optimal QoE, so content royalties are a cost of doing business. For FAST services and others working with 8-bit 1080p content, HEVC and AV1 are largely an efficiency play, and content royalties radically change the break-even math.
Nokia and Amazon Settle Worldwide Claims, Including for Content Royalties
In early 2025, Nokia and Amazon reached a global patent settlement after Nokia sued Amazon over video and streaming-related patents across multiple jurisdictions, including Germany, the U.S., and the U.K., and in the European Unified Patent Court. This was one of the first known settlements relating to content royalties as opposed to device royalties.
In its complaint, Nokia “accused Amazon of allowing customers of its Prime Video, Freevee, and Twitch services to stream video content encoded in the H.264 standard and a newer one called H.265,” alleging infringement of at least 15 Nokia patents. Nokia has described these as “multimedia inventions” used in Amazon’s streaming services and devices, not just chipset IP (see Figure 1).
Figure 1. Summary of Nokia’s claims against Amazon
The threat of German injunctions and related pressure from the U.S. International Trade Commission were particularly notable because they created real, near-term remedies against both the Prime Video service and Fire TV devices, rather than abstract legal threats.
Amazon was the first tier-1 streaming service forced to the table over video and streaming patents, not just hardware licensing. Nokia signed at least two other large agreements without naming names. Inside legal and commercial teams, the message translated quickly and bluntly: If Amazon can be enjoined over streaming functionality, no major service is immune. The longstanding strategy of ignoring codec and streaming IP exposure and assuming it would never block service delivery stopped looking credible.
Nokia is not the only codec IP owner asserting rights over streamed content. In early 2025, InterDigital sued Disney in the U.S., Brazil, Germany, and the European Unified Patent Court over technologies used in Disney+, Hulu, and ESPN+. The Munich (Germany) Regional Court ruled that Disney cannot use InterDigital’s advanced video compression technology on its streaming platforms in Germany, issuing an injunction based on infringing EP 1 905 233, which covers dynamic overlay of one video stream with another (for example, subtitles). A Brazil court found in favor of InterDigital, while the suits in the U.S. and European Unified Patent Court are still pending.
Implications for Publishers
Broadcom has pursued Netflix in the U.S. and Europe over content encoded using HEVC, including a German case in which Netflix was initially found to infringe a Broadcom streaming patent before a second court found the Broadcom patent to be invalid. Not to cut too fine a slice, but while this relieves Netflix of any liability to Broadcom, it does not reverse the broader conclusion that encoding and delivering content using a codec can infringe patents covering that codec’s technology.
The bottom line is that beyond content-side patent pools, the largest streaming services should expect direct approaches and potential enforcement from Nokia, InterDigital, Broadcom, and other patent owners, not only for HEVC or AV1 but potentially for H.264 and related streaming functionality as well. Once a few of these suits are settled, it’s likely that all of these patent owners will target progressively smaller services.
Although I do have a CPA in my distant background, I’m not here to offer financial advice. Still, by 2026, no responsible financial executive should greenlight the deployment of any advanced video codec without at least considering a reserve for potential royalty obligations. With device-side pools operating, content-side licensing programs publishing rates, and enforcement proven possible at scale, a zero-royalty assumption increasingly looks optimistic rather than conservative.
Once finance starts booking reserves or accruals for codec royalties, the psychological battle is over; those costs are 99% of the way toward normalization. At that point, the question stops being “Can we avoid paying for codecs?” and becomes “Which codecs are worth paying for and at what scale?”
Sisvel Claims to Have Licensed Half The AV1 Device Market
On July 31, 2025, Sisvel announced that its AV1 patent pool had licensed roughly 50% of the finished-product AV1 device market, covering TVs, set-top boxes, and other consumer hardware shipping with AV1 decode. By any measure, 50% is quite an accomplishment, but particularly so for a technology launched as “royalty-free.”
In a more general sense, this announcement makes it increasingly difficult to assert that AV1 is somehow different than HEVC or VVC and is royalty-free. That said, to quote Yogi Berra, “It ain’t over till it’s over.” H.264 and HEVC are done and dusted, with multiple court decisions for devices and content and Amazon Prime paying content-related royalties to Nokia. As far as I know, there are no equivalent cases for AV1 on either the device or the content side and no content-side agreements for AV1 like the Nokia-Amazon Prime agreement.
All that said, we won’t know if AV1 is truly royalty-free until it’s litigated in court in all relevant jurisdictions. However, from a financial perspective, if you’re a company that’s large enough to meet one of Access Advance’s minimum royalty thresholds, anticipating that AV1 will be royalty-free has gone from prudent planning to wishful thinking.
InterDigital Acquires Deep Render
The first thing any publisher should know about InterDigital’s acquisition of Deep Render (see Figure 2) is that it will not matter operationally for years. Nothing about this deal changes encoding ladders, device support, or delivery economics in the current planning cycle. Whatever emerges from AI-native compression research will arrive closer to 2030 than 2026.
Figure 2. InterDigital locks up patent IP for AI-based codecs.
The reason the acquisition belongs in a 2025 codec year in review is directional. If you were hoping that AI-based codecs would disrupt the current codec licensing schema, this deal signaled “meet the new boss, same as the old boss.”
By way of background, InterDigital is a non-practicing entity (NPE) that drives “the development of advanced technologies and contributes them to global standards that define the future of wireless communication and video.” The NPE designation is descriptive, not derogatory. If you’re looking for validation of InterDigital’s technology contributions or business model, look no further than an $8 billion market cap from around the start of 2025.
InterDigital didn’t buy Deep Render to release a new AI-based codec and sweep the feet from under MPEG and the International Telecommunication Union (ITU). Rather, according to Rajesh Pankaj, InterDigital’s CTO, “Acquiring Deep Render means that we’re perfectly positioned to lead the development of the next generations of video technologies, building on our existing leadership in HEVC and VVC.” This explicitly ties AI-native compression to the same standards-driven, IP-centric model that has defined previous codec generations.
For publishers, the implication is economic rather than technical. If AI-native compression techniques are being accumulated by the same organizations that monetized AVC, HEVC, and VVC, then the post-AV2 world won’t reset the royalty conversation. Instead, it points toward a future where new codecs will arrive with familiar IP exposure. The acquisition does not make AI codecs imminent, but it makes the longterm direction difficult to ignore.
Brazil’s TV 3.0 Decree Locks In VVC + LCEVC
On Aug. 27, 2025, Brazil’s president formally established the ATSC 3.0-based DTV+ (“TV 3.0”) system as the country’s next-generation free-to-air broadcast standard, with commercial services expected to start in time for the 2026 World Cup. VVC and LCEVC vendors have pointed to Brazil’s TV 3.0 plan, which describes VVC as the main video codec and LCEVC as an enhancement layer, as a beachhead that will drive global adoption. A review of what the other major standards families are doing in the same time frame suggests a more fragmented reality.
Let’s start with ATSC, which applies to the U.S., South Korea, and other, smaller jurisdictions. In 2024, ATSC introduced VVC as a Candidate Standard and stressed that it would be “an option in addition to” HEVC, not a replacement. In July 2025, ATSC approved VVC as a full ATSC 3.0 standard, but made it optional, while leaving HEVC as the practical baseline for advanced codecs. DVB defines broadcast specs across Europe and other regions, and its 2022–2023 updates added both VVC and China’s AVS3 to its core DVB-AVC specification. However, DVB explicitly describes its role as “setting standards, not adoption,” meaning that actual deployments of advanced codecs are up to the individual countries.
In China, AVS3 is the nationally backed next-generation codec, and it is already used in 4K/8K broadcasts. DVB’s inclusion of AVS3 alongside VVC and AV1 reflects broader awareness of AVS3 beyond China.
Some have claimed that VVC adoption will be driven by Chinese VVC patent owners, especially those behind video-heavy services like TikTok/Douyin and other major platforms. However, there is no clear indication that large Chinese streaming platforms are pushing VVC as a commercial standard in the same way YouTube, Facebook, and Netflix have driven AV1 adoption. In China today, VVC interest appears more rooted in standards participation and patent involvement than in overt market or publisher pressure to deploy the codec at scale.
These standards are summarized at a high level in Table 2, although the actual facts are so detailed that a standards wonk could disagree with almost every entry. The bottom line as it relates to VVC and LCEVC is that Brazil is a standalone jurisdiction, and what happens in Brazil likely stays in Brazil. Yes, at some point in the not-so-distant future, many TVs in Brazil will have VVC/LCEVC decode, but that influence is likely local, not global.
Table 2. Codec usage in broadcast standards
As we’ve seen, ATSC and DVB adopt and incorporate codecs, but don’t dictate their use. As I’ll cover in a moment, AV1 has much greater publisher adoption than VVC and LCEVC. That’s why AV1 hardware decode is included in most currently shipping TVs, while VVC and LCEVC decode, which are available in multiple chipsets, are rare, if not totally unavailable. In this regard, Brazil appears to be more of an Alamotype scenario than a Normandy.
AV2 Arrives, AV1 Advances
In late 2025, AOMedia finalized the AV2 specification and bitstream. This was not a deployment event, but it was a real milestone. AV2 moved from research slides
and conference talks into something silicon vendors and encoder developers could target. From that point on, AV2 belonged in the 5- to 10-year road map section of any serious codec strategy deck, with potential early deployments from the big three AV1 users: Meta, Netflix, and YouTube (see Figure 3).
Figure 3. Netflix’s timeline of AV1 adoption and deployment (image credit: Netflix)
Here’s what we know about AV2 from recent AOMedia presentations by Netflix and Meta. Architecturally, the codec remains a conventional hybrid block-based design, which simplifies hardware planning and continuity with existing pipelines. Film Grain Synthesis is retained as a mandatory tool, with improvements underway, reinforcing AV2’s suitability for premium and cinematic content.
Regarding encoding efficiency, measured gains over AV1 are consistently in the high-20%-to-mid-30% range across random access, adaptive streaming, HDR, extended color formats, and subjective testing. From a purely compression standpoint, AV2 appears capable of closing much of the efficiency gap between AV1 and VVC, although not clearly surpassing it.
What remains unknown is just as important. No concrete data was shared on encoder complexity, software decode performance, or power efficiency, all of which will heavily influence real-world deployment. Hardware feasibility has been reviewed, but no silicon timelines or decoder benchmarks exist yet.
Licensing and IP exposure were not discussed, despite the codec’s reliance on long-established compression architectures. Sisvel has already announced its intention to create an AV2-related patent pool, and you would expect the same from Avanci and Access Advance.
What AV2 Revealed About AV1
Ironically, AV2’s arrival helped clarify how the AV1 ecosystem had progressed to enable deployments beyond AV1’s largest supporters: Meta, Netflix, and YouTube. Netflix recently reported that AV1 was powering roughly 30% of its streams and had expanded into HDR delivery, finally providing feature parity with HEVC, if not the proven installed base of 4K/ HDR-compatible players.
Regarding playback in the living room, Netflix also reported, “Over the past five years (2021–2025), 88% of large-screen devices, including TVs, set-top boxes, and streaming sticks, submitted for Netflix certification have supported AV1, with the vast majority offering full 4K@60fps capability.” In the living room, where premium services deliver most of their content, AV1 has certainly reached critical mass.
With AV1 playback support available in browsers almost since Day 1, this left mobile as the last remaining market where AV1 hardware playback support lags significantly behind HEVC. Here, unlike the every-company-for-itself VVC camp, AOMedia funded the development of the highly optimized dav1d software decoder, enabling efficient AV1 playback on Android devices without hardware decode.
Netflix initiated its AV1 support on Android via dav1d, a move Meta later followed with Reels. However, YouTube was criticized when it began defaulting to dav1d-based software playback on older and mid-range phones, with users reporting higher CPU load, dropped frames, and noticeable battery drain during YouTube playback. While this had all the feel of a mountain-out-of-a-molehill social media exaggeration, it may discourage other third-party streaming publishers from experimenting with software-based playback.
Mobile aside, AV1 has clearly achieved sufficient playback compatibility to enable publishers beyond the big three to consider deploying it. Some, like Evolution Gaming, have done so already. We’ll learn in 2026 whether this triggers a trickle or a torrent of additional deployments.
Honorable Mentions
There are two other noteworthy events that don’t make the main story because of their lack of immediate impact on publishers.
First, Access Advance’s acquisition of Via Licensing Alliance’s (Via LA) HEVC/VVC program combines the two main HEVC/VVC patent pools under a single administrator and rebrands the Via LA program as VCL Advance (Video Codec Licensing Advance). This gives licensees a one-stop shop for most HEVC and VVC standard-essential patents. For TV, set-top box, and handset vendors, this promises cleaner paperwork, fewer overlapping contracts, and the potential to leverage Access Advance’s existing multi-codec and bridging structures across HEVC and VVC portfolios. However, as Via LA didn’t have a content pool, this transaction involves encoding/decoding royalties rather than content. This makes it relevant to the market
in general, but not particularly so for publishers. Second, on July 14, JVET released its Proposed Timeline and Requirements for the Next-Generation Video Coding Standard document, formally setting a target of at least 40% bitrate reduction over VVC Main 10 for 4K-and-above content at similar subjective quality. This effectively signified the official kickoff of the H.267 effort inside the standards bodies. Still, given that standardization is not expected until around 2028 and real-world deployments are likely a decade away, it remains far more critical to standards engineers than to streaming publishers planning deployments this decade.
Third, Google’s Android 17 adds native support for Versatile Video Coding (VVC/H.266) on devices with compatible hardware decoders, putting VVC on equal technical footing with HEVC and AV1 inside the Android media stack. To be clear, Google hasn’t added a software VVC player to Android. It’s just made it easier to access VVC hardware decode that might exist on the mobile device or TV.
Given that VVC hardware support is generously called nascent on these platforms, it’s probably an overstatement to call this an “endorsement” of VVC by Google. Still, for a codec that hasn’t made much visible progress over the last few years, it’s a welcome sign from a company that’s otherwise firmly in the AV1/AV2 camp.
2025 In Review
If I had to describe 2025 in an elevator ride, it would be the year that content royalties became real, AV1 achieved critical mass, and VVC somehow shrank from an already-low profile. And the year that assuming AV1 is “royalty-free” became financially irresponsible.
Related Articles
HEVC rode the 4K/HDR wave to success, but AV1, VVC, and LCEVC lack an equivalent killer app. With CDNs racing to the bottom on pricing and new royalty pools threatening to increase costs, codec adoption is no longer just about technical merit—it's about survival. These are the issues I'll explore in this article.
28 Mar 2025
It's been 3-plus years since the MPEG codec explosion that brought us VVC, LCEVC, and EVC. Rather than breathlessly trumpet every single-digit quality improvement or design win, I'll quickly get you up-to-speed on quality, playability, and usage of the most commonly utilized video codecs and then explore new codec-related advancements in business and technology.
04 Apr 2024
Welcome to your codec update for 2023. I'm here to help you decide whether it's time to go all in on AV1, VVC, LCEVC, or EVC or whether it's better to stick with H.264, VP9, and HEVC. Along the way, I'll include 2022-2023 updates to identify significant changes that have occurred since last year.
12 Apr 2023