Apple shipped the M1 in late 2020 and stopped selling Intel Macs years ago. So the obvious assumption is that developers finished the move to Apple Silicon a while back. The shipped binaries say otherwise.
Across the 10,662 Mac apps we track at the binary level, 16.1% still ship Intel-only, with no native arm64 slice at all. That is roughly one in six, more than five years after the M1, still depending entirely on Rosetta 2 to run on every Mac sold today.
If you size platform transitions for a living, that share is the number that matters. It is the part of the migration that hasn't happened, measured from what apps actually ship rather than what their release notes claim.
We read the architecture from the binary, not the badge
Most "Apple Silicon readiness" claims trace back to press releases, store badges, or developer surveys. We don't ask. We read the Mach-O binary inside each shipped app bundle and record which CPU architectures it carries: Intel (x86_64), Apple Silicon (arm64), or both in a single Universal 2 binary.
That distinction is invisible to a source-repo scanner. A repo can target both architectures while the artifact a user downloads carries only one. We look at the slice that actually ships, the thing that runs on the customer's machine, which is why our split can differ from a vendor's marketing posture.
Most Mac apps already ship native
Here is the full split:
- Universal (Intel + ARM): 57.2%. Runs native on both. The compatibility-maximizing default, and where most of the apps we track sit today.
- Apple-Silicon-only (arm64): 26.7%. Native Apple Silicon, no Intel slice at all.
- Intel-only: 16.1%. No arm64 slice, Rosetta 2 or nothing.
Add the first two together and more than eight in ten Mac apps ship a native Apple Silicon binary of some kind. The transition is the majority position by a wide margin. But "mostly done" is not "done," and the remaining sliver is the interesting part.
The holdouts are the software that has gone quiet
Five-plus years in, an Intel-only slice is less a transitional state than a signal.
We don't claim to know each holdout's reason from the binary alone, so here is what we measure versus what we can only infer. What we measure is the architecture slice. What we can reasonably infer about the kind of software that lags is shaped by a couple of structural facts:
- Native toolchains have had first-class arm64 support for years. The apps least likely to have rebuilt are the older or less actively maintained ones, where nobody has shipped a fresh binary since the Intel era.
- Abandonment looks identical to inertia in the binary. An Intel-only slice doesn't tell you whether the vendor chose Intel-only or has not shipped since 2020. Either way, the end user is running under Rosetta 2.
For an analyst, the point is that the long tail of the Mac migration is real and quantifiable. It is a sixth of the corpus, and the clearest available proxy for software that has stopped moving.
arm64-only is the number worth watching
The forward indicator isn't the Universal share. It's the arm64-only share, currently just over a quarter of the corpus.
Universal is the conservative default: ship both, break nothing. Dropping the Intel slice entirely is a deliberate decision, a developer saying Intel Macs are no longer worth supporting in a new build. As the arm64-only share grows at the expense of the Universal cohort, you are watching how ready developers are to treat Intel Macs as a rounding error.
For a dev-tools company, this is a roadmap input. The arm64-only share tells you how aggressively your customers will drop Intel, which tells you when you can stop testing and building an x86_64 path without alienating them. Right now, with well over half still hedging with Universal, the answer is "not yet." Track that share across snapshots and you have a defensible trigger for the decision rather than a guess.
Minimum-OS targets are a second view of the same question
Architecture is one axis of how modern an app is. The minimum macOS version it targets is another. The top minimum-OS cohorts among the Mac apps we track:
- 10.13 High Sierra: 13.9%, the single largest cohort.
- 12.0 Monterey: 12.1%.
- 11.0 Big Sur: 10%.
- 10.15 Catalina: 9.3%.
- 14.0 Sonoma: 5.2%.
- 13.0 Ventura: 4.9%.
There is a long tail back to 10.11. The largest single cohort still names High Sierra as its floor, a release that predates Apple Silicon. A minimum that low can mean an old toolchain, or it can be a deliberate choice to keep supporting old hardware. The binary alone doesn't say which, so it is worth checking against the Intel-only set rather than assuming the two are the same apps.
Big Sur is the first macOS that ran on Apple Silicon, so the cohorts at Big Sur and later are the Apple-Silicon-era releases. But a modern minimum-OS target doesn't force a modern architecture. Monterey and later still run on Intel Macs, so an app can require a recent macOS and still ship Intel-only. Minimum-OS and architecture are two related views of the same question, not the same measurement. For what these apps are built with in the first place, see what these apps are built with, which covers the same corpus's runtime and toolchain split.
How far off "finished" is
A finished Apple Silicon transition, read from binaries, would be a corpus with essentially no Intel-only apps: every shipped artifact either Universal or arm64-only, and the arm64-only share climbing as developers retire Intel support.
We are not there. A meaningful share still ships Intel-only, and the Universal majority shows most developers still aren't ready to commit to arm64-only. The leading edge is the direction of travel, but it is not yet the center of gravity.
That gap is the opening for anyone whose business depends on knowing where the platform actually stands. This isn't a "done or not done" event. It is a distribution that moves snapshot over snapshot, measurable today from the one source that can't be spun: the binary that ships to the user.
About the data
Figures are from Desktop Insights' analysis of more than 25,000 macOS and Windows desktop apps, June 2026 snapshot. The architecture split covers 10,662 Mac apps that carried a detected CPU architecture. Architecture and minimum-OS targets are read from the shipped binary, what end users actually run, not from source repositories.
Track it yourself. You can follow Apple Silicon readiness across any app set in the Desktop Insights directory, or ask us for the architecture dataset to benchmark platform-transition progress for your own market.