Firefox 153 will add Vulkan Video decoding as Mozilla widens Linux GPU support
The initial code path, credited to NVIDIA and Red Hat engineers, targets a long-running hardware video gap in Firefox on Linux.
By Ryan Merket · Published
Why it matters
Firefox's Vulkan Video work is a quiet infrastructure bet: Mozilla is reducing dependence on VA-API, a Linux video path that has left NVIDIA and embedded GPU users relying on workarounds.
Mozilla has merged initial Vulkan Video decoding support for Firefox, work Phoronix reported was led by NVIDIA engineer Tymur Boiko and Red Hat's Martin Stransky, putting a new GPU-accelerated video path on track for Firefox 153 in July.
The important word is "initial." Phoronix says Firefox 153.0 is expected on July 21, 2026, assuming no last-minute issues. The report does not establish which codecs, GPUs, drivers, distributions, or operating systems will be covered at launch, or whether Vulkan Video decoding will be enabled by default. What Mozilla has landed is a second route around a Linux video stack that has never been uniform.
The work attacks a Linux-specific bottleneck
Firefox on Linux has long centered hardware video decoding around VA-API, the Video Acceleration API. That has worked unevenly because Linux graphics drivers do not all expose the same acceleration interfaces. Phoronix specifically points to NVIDIA users, who have relied on projects such as NVIDIA-VAAPI-Driver to map Firefox's VA-API path onto NVIDIA's NVDEC interfaces.
That translation layer is the tell. When a browser needs an adapter project to reach a major GPU vendor's video engine, the abstraction is not doing the whole job. Phoronix also notes that smaller Arm and embedded graphics drivers have been largely outside the VA-API path, which matters as Linux keeps expanding beyond traditional desktop PCs into handhelds, low-power devices and bespoke hardware.
Vulkan Video gives Firefox another decode target in a graphics API already designed to span vendors and platforms. That does not make VA-API obsolete inside Firefox, and the source material does not say Mozilla is replacing it. It does show Mozilla accepting that one Linux acceleration path has left too many edge cases for too long.
The maintainers, not the marketing team, moved this
This is not a feature announcement packaged for consumers. The trail runs through Mozilla's public development machinery: Phoronix says a Bugzilla report opened about three months earlier over Firefox's lack of Vulkan Video support has been closed, and points to recent Firefox Git activity that culminated during the week ending June 7, 2026.
The named contributors are the useful signal. Boiko works at NVIDIA, the vendor whose Linux users have had one of the more awkward Firefox acceleration stories. Stransky works at Red Hat, whose engineers have long been part of the Linux desktop plumbing around Firefox, graphics and system integration. Mozilla benefits from the patch, but the incentives are broader: NVIDIA wants its hardware video engines reached through modern Linux software paths, and Red Hat wants the open desktop stack to depend on fewer vendor-specific workarounds.
That is how browser infrastructure often moves. A user-visible improvement begins as maintenance work across a bug tracker, a graphics API, a driver stack and a release train. By the time it reaches stable Firefox, the most important decision has already been made: which platform abstraction Mozilla is willing to support in the browser core.
Firefox 153 is the real test
The July 21 target matters because it puts the work on a near-term stable release path, not just an experimental branch. But the source material leaves the operational questions unanswered. There is no confirmed support matrix for codecs. There is no stated minimum Vulkan driver version. There is no confirmed default setting. There is no performance claim to compare against VA-API or software decoding.
Those omissions are not minor. Hardware video decoding is only valuable when it reliably reduces CPU load, battery drain and thermals on real machines. A code path that works on one vendor's current drivers can still disappoint users if distributions ship older stacks, if codec support is narrow, or if the feature remains behind a preference.
Still, the direction is clear. Mozilla is adding Vulkan Video because the old Linux answer was incomplete. Firefox 153 will show whether the first implementation is a narrow foothold or the start of a more portable hardware decode layer for the browser.