Nearly a decade after the last major release, NetBSD is on the verge of shipping version 11.0. The third release candidate dropped in late June 2025, and it signals that one of the oldest open-source operating systems still standing is finally ready to close a chapter that has tested the patience of even its most loyal users.
The road here has been long. Extraordinarily long.
NetBSD 10.0 shipped in March 2024 after what many in the BSD community considered an agonizing gestation period. Now, barely a year later, the project is pushing toward 11.0 with a pace that looks almost brisk by comparison. Release candidate 3, as reported by Phoronix, arrived with a focused set of fixes and improvements that suggest the development team is polishing rather than overhauling — exactly what you want to see at the RC3 stage.
What NetBSD 11.0 RC3 Actually Brings
The latest candidate builds on the substantial foundation laid in earlier pre-release versions. NetBSD 11.0 brings a modernized kernel, updated hardware support, and improvements across its networking stack and file systems. The release includes updates to the toolchain, with newer versions of GCC and LLVM/Clang available for compilation. There are also improvements to the project’s hallmark portability layer — NetBSD still supports an almost absurd number of hardware architectures, from modern x86-64 and ARM64 systems down to vintage platforms that most operating systems abandoned years ago.
RC3 specifically addresses bugs identified during the RC1 and RC2 testing cycles. This is a stabilization release, not a feature dump. Bug fixes in device drivers, corrections to edge-case behavior in the networking code, and minor userland improvements make up the bulk of the changes. The release engineering team has been methodical about triaging regressions, and the shrinking diff between release candidates suggests the final 11.0 release is imminent.
For those tracking the broader BSD family, this matters. NetBSD has historically served as an upstream source for ideas and code that eventually find their way into FreeBSD, OpenBSD, and even commercial products. Its pkgsrc package management system is used across multiple operating systems, including macOS and Linux in some deployments. A healthy NetBSD release cycle benefits more than just NetBSD users.
But let’s be direct about the project’s position. NetBSD doesn’t command the server market share of FreeBSD or the security-focused mindshare of OpenBSD. It occupies a niche — portability, clean design, research-friendly licensing — that doesn’t generate headlines. What it does generate is infrastructure that other projects depend on, often without acknowledgment.
The Broader BSD Moment
NetBSD 11.0’s approach to release comes at an interesting time for BSD-derived operating systems generally. FreeBSD 14.x continues to mature, with version 14.2 released earlier this year bringing performance improvements and updated drivers. OpenBSD maintains its steady six-month release cadence under Theo de Raadt’s uncompromising leadership, with version 7.7 expected later in 2025. And DragonFly BSD, the smallest of the major BSDs, continues its own development trajectory focused on the HAMMER2 file system.
The competitive pressure isn’t really between BSDs, though. It’s between BSD and Linux.
Linux dominates servers, cloud infrastructure, embedded devices, and increasingly the desktop through distributions like Ubuntu, Fedora, and Arch. The BSDs collectively represent a fraction of deployed systems. Yet they persist — and in some areas thrive — because they offer things Linux doesn’t. A single coherent base system rather than a loose assembly of packages. Licensing under BSD or ISC terms rather than the GPL, which matters enormously to companies embedding operating systems in commercial products. And in NetBSD’s case, a degree of architectural portability that no other general-purpose OS matches.
Sony’s PlayStation operating systems have historically drawn from FreeBSD. Nintendo’s Switch uses components traceable to BSD heritage. Juniper’s router firmware runs on a BSD base. These aren’t hobbyist deployments.
NetBSD’s own deployment stories tend to be quieter. Embedded systems. Research projects at universities. Network appliances from vendors who prefer BSD licensing for proprietary products. The kind of work that doesn’t show up in Stack Overflow surveys but keeps the internet’s plumbing functioning.
So what does RC3 mean practically? For system administrators and developers who’ve been testing RC1 and RC2, it means another round of validation before committing to production deployments. For the NetBSD release engineering team, it means they’re close enough to the finish line that a final release in the coming weeks looks realistic. And for the BSD community at large, it’s a sign that the project’s development velocity — which had been a source of genuine concern during the long gap between NetBSD 9.x and 10.0 — has meaningfully improved.
The project’s release notes detail dozens of specific changes across the kernel, userland utilities, third-party software updates, and documentation fixes. Among the more notable improvements in the 11.0 branch are better support for modern ARM hardware, including various single-board computers popular with embedded developers. There’s also improved support for newer Intel and AMD processors, updated Wi-Fi drivers, and enhancements to the ZFS implementation that NetBSD has been integrating.
One area where NetBSD continues to differentiate itself is in its approach to the base system. Unlike Linux distributions, which assemble hundreds of independent projects into a coherent whole, NetBSD develops its kernel and core userland as a unified project. This means the entire base system is covered by a consistent license, built with a consistent toolchain, and maintained by a single group of developers who understand the interactions between components. It’s a philosophy shared with the other BSDs but taken perhaps furthest by NetBSD, which has historically been the most willing to write its own implementations of standard tools rather than importing them from external projects.
That philosophy has costs. Fewer developers means slower progress on some fronts. Features that arrive in Linux quickly — like certain hardware drivers or container technologies — take longer to reach NetBSD, if they arrive at all. The project doesn’t have corporate sponsors on the scale of Red Hat backing Fedora or Canonical backing Ubuntu. The NetBSD Foundation relies on donations and volunteer labor.
Yet the project endures. It’s been active since 1993, making it one of the longest-running open-source projects in existence. Thirty-two years. Through the dot-com boom, the Linux ascendancy, the cloud computing wave, and now the AI infrastructure buildout, NetBSD has kept shipping code.
RC3 is not the finale. But it’s close enough that anyone with a stake in BSD systems — whether as a user, developer, or vendor — should be paying attention. The final 11.0 release, when it arrives, will represent the project’s most rapid major-version turnaround in years, and a signal that NetBSD’s development community has found a sustainable rhythm after a period of uncertainty.
For those who want to test RC3, images are available for dozens of platforms from the NetBSD project’s official mirrors. Bug reports, as always, are welcome. The project needs them. That’s how open source works — not through press releases, but through people running code, breaking things, and filing reports so the next build is better than the last.
WebProNews is an iEntry Publication