Apple Neural Engine: Architecture, Programming, and Performance
2026/06/21 by Spencer H. Bryngelson · 11 voices
#cs.AR #cs.OS #cs.PF
paper · pdf
Abstract
The Apple Neural Engine (ANE) is the fixed-function matrix accelerator that has shipped in Apple systems-on-chip since the A11-class iPhone and iPad chips and the M1-class Mac chips, exposed to applications only through the Core ML model framework. This guide reports a reverse-engineered account of the engine, based on direct measurement on Apple silicon and static analysis of the private runtime, compiler, kernel driver, and firmware. It documents the datapath and the roofline that bound the engine's throughput and energy, the dispatch route that reaches it below Core ML, the compiler and on-disk program format, the weight-compression scheme, and the kernel driver, firmware, and command protocol beneath them. The account covers the A11 through A18 and M1 through M5 families, with per-chip target tables and an operation-by-device matrix; the direct measurements are on the M1 and M5. Claims are labeled as measured, decompile-derived, or predicted, and the methodology and open questions are recorded. The direct route is callable from ordinary user space but remains undocumented, unsupported, and version-fragile; it is intended for measurement, research, and on-device work, not for shipping software, where Core ML remains the supported path.
Discussions
- Apple Neural Engine: Architecture, Programming, and Performance [hn, 234 points, 29 comments]
- Holy shit! A three-hundred-page paper about the Apple Neural Engine. arxiv.org/abs/2606.22283 [bsky, 68 points, 1 comments]
- Mentioned a few times I've been looking at ANE (Apple Neural Engine, the fp16 chip in all new Macs & iPhones). Just came across another great resource, published a few weeks ago, reverse engineered a [bsky, 16 points, 1 comments]
- Apple Neural Engine: Architecture, Programming, and Performance [lobsters, 5 points, 0 comments]
- [16/30] 249 Upvotes, 29 Comments, 2 Posts, arXiv:2606.22283 🆕Apple Neural Engine: Architecture, Programming, and Performance Spencer H. Bryngelson [bsky, 1 points, 1 comments]
- Apple Neural Engine: Architecture, Programming, and Performance https://lobste.rs/s/6cdrev #hardware #performance [bsky, 1 points, 0 comments]
- Apple Neural Engine: Architecture, Programming, and Performance https://arxiv.org/abs/2606.22283 (https://news.ycombinator.com/item?id=48702825) [bsky, 0 points, 0 comments]
- Apple Neural Engine: Architecture, Programming, and Performance https://arxiv.org/abs/2606.22283 (http://news.ycombinator.com/item?id=48702825) [bsky, 0 points, 0 comments]
- Apple Neural Engine: Architecture, Programming, and Performance https://arxiv.org/abs/2606.22283 [bsky, 0 points, 0 comments]
- Apple Neural Engine: Architecture, Programming, and Performance https:// arxiv.org/abs/2606.22283 # apple # arxiv [mastodon, 0 points, 0 comments]
- Apple Neural Engine: Architecture, Programming, and Performance https://arxiv.org/abs/2606.22283 https://news.ycombinator.com/item?id=48702825 [bsky, 0 points, 0 comments]
Related