{"slug": "apple-neural-engine-architecture-programming-and-performance", "title": "Apple Neural Engine: Architecture, Programming, and Performance", "summary": "Researchers reverse-engineered the Apple Neural Engine (ANE), documenting its architecture, programming interface, and performance across A11-A18 and M1-M5 chips. The study reveals a direct user-space route to the ANE, though it remains undocumented and version-fragile, with Core ML as the supported path for shipping software.", "body_md": "# Computer Science > Hardware Architecture\n\n[Submitted on 21 Jun 2026]\n\n# Title:Apple Neural Engine: Architecture, Programming, and Performance\n\n[View PDF](/pdf/2606.22283)\n\nAbstract: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.\n\n## Submission history\n\nFrom: Spencer Bryngelson [[view email](/show-email/cf3990d2/2606.22283)]\n\n**[v1]** Sun, 21 Jun 2026 00:17:34 UTC (407 KB)\n\n### Current browse context:\n\ncs.AR\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/apple-neural-engine-architecture-programming-and-performance", "canonical_source": "https://arxiv.org/abs/2606.22283", "published_at": "2026-06-27 23:30:14+00:00", "updated_at": "2026-06-28 00:05:01.946919+00:00", "lang": "en", "topics": ["ai-chips", "machine-learning", "neural-networks"], "entities": ["Apple", "Apple Neural Engine", "Core ML", "M1", "M5", "A11", "A18", "Spencer Bryngelson"], "alternates": {"html": "https://wpnews.pro/news/apple-neural-engine-architecture-programming-and-performance", "markdown": "https://wpnews.pro/news/apple-neural-engine-architecture-programming-and-performance.md", "text": "https://wpnews.pro/news/apple-neural-engine-architecture-programming-and-performance.txt", "jsonld": "https://wpnews.pro/news/apple-neural-engine-architecture-programming-and-performance.jsonld"}}