12 comments

  • GeekyBear30 minutes ago
    The recent news is that Apple is supposedly replacing the Core ML framework with an updated version that will make it easier to integrate third party LLMs into your apps.<p>&gt; the company is also planning a few other software-based AI upgrades, including a new framework called Core AI. The idea is to replace the long-existing Core ML with something a bit more modern.<p><a href="https:&#x2F;&#x2F;www.bloomberg.com&#x2F;news&#x2F;newsletters&#x2F;2026-03-01&#x2F;apple-touch-screen-macbook-pro-fall-2026-details-cheap-macbook-launch-core-ai-mm7rcg48" rel="nofollow">https:&#x2F;&#x2F;www.bloomberg.com&#x2F;news&#x2F;newsletters&#x2F;2026-03-01&#x2F;apple-...</a>
  • LatencyKills1 hour ago
    I worked on the Xcode team for years and know the lengths Apple goes to make this stuff difficult to figure out.<p>I just wanted to say that you’ve done an excellent job and am looking forward to the 3rd installment.
  • Octoth0rpe1 hour ago
    Part 2 has benchmarks: <a href="https:&#x2F;&#x2F;maderix.substack.com&#x2F;p&#x2F;inside-the-m4-apple-neural-engine-615?r=1afltp" rel="nofollow">https:&#x2F;&#x2F;maderix.substack.com&#x2F;p&#x2F;inside-the-m4-apple-neural-en...</a><p>6.6 FLOPS&#x2F;W, plus the ability to completely turn off when not in use, so 0W at idle.
  • love2read2 hours ago
    This article was clearly written by a human (and AI) but still has a few &quot;LLMisms&quot; such as:<p>- The key insight - [CoreML] doesn&#x27;t XXX. It YYY.<p>With that being said, this is a highly informative article that I enjoyed thoroughly! :)<p>The article links to their own Github repo: <a href="https:&#x2F;&#x2F;github.com&#x2F;maderix&#x2F;ANE" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;maderix&#x2F;ANE</a>
    • rafram14 minutes ago
      Also the Prior Art section, which has telltale repetition of useless verbs like &quot;documenting,&quot; &quot;providing insight into,&quot; and &quot;confirming&quot; on each line. This was definitely AI-written, at least in part.
    • walthamstow1 hour ago
      We&#x27;ve got about a year before so many people are interacting with LLMs on a daily basis that its style starts to reverse infect human speech and writing
      • pixl971 hour ago
        This said, there were people that talked like this before LLMs, it didn&#x27;t develop this whole cloth.
        • DrScientist25 minutes ago
          Exactly. LLM&#x27;s are mimics.<p>People seem to be going around pointing out that people talk like parrots, when in reality it&#x27;s parrots talk like people.
      • Angostura1 hour ago
        My honest take? You&#x27;re probably right
        • sholladay33 minutes ago
          You are absolutely right.<p>Here is why you are correct:<p>- I see what you did there.<p>- You are always right.
  • behnamoh26 minutes ago
    It&#x27;s insane that the source code of ANE is not available <i>even to the MLX team</i>, possibly one of the reasons Awni (MLX project head) left Apple.
    • mathisfun12323 minutes ago
      Tell me you&#x27;ve never worked at a hardware company without telling me lololol
      • behnamoh22 minutes ago
        Yes I haven&#x27;t worked at a hardware company, nothing to be ashamed of!
  • mattlangston2 hours ago
    The future is bright for software engineers.<p>The big takeaway isn&#x27;t reverse engineering the ANE per se, but what Manjeet could do with his software engineering skills when accelerated by AI.<p>This is a good example of the present state of software engineering. Not future state - present state.
  • daoistmonk39 minutes ago
    Tangential: Is anyone doing something similar to accelerate the support matrix of Linux on anything higher than M2?
  • kamranjon1 hour ago
    I have always wondered if the neural engine could be used for training - pretty excited for part 3 of this to see if the juice is actually worth the squeeze
  • FL33TW00D3 minutes ago
    Unreadable Claude slop
  • techpulse_x1 hour ago
    [dead]
  • poszlem3 hours ago
    Genuine question, not trying to throw a shade or anything, but are those cores actually useful with the state of apple intelligence being what it is?
    • rahkiin2 hours ago
      They are also used by ML models that are deeply integrated in macos and ios without you knowing. Like object and text detection in images.
      • willis9361 hour ago
        I wish they would (or wouldn&#x27;t if they are) hook it up to the ios keyboard.
      • geerlingguy1 hour ago
        And help in Photos, Final Cut Pro, and other apps.
    • dagmx1 hour ago
      If you strip away the branding, Apple has and continues to ship a ton of algorithms that likely use the ANE and end users can use CoreML to do the same.<p>Just some things that people will likely take for granted that IIRC Apple have said use the ANE or at least would likely benefit from it: object recognition, subject extraction from images and video, content analysis, ARKit, spam detection, audio transcription.
      • sroussey29 minutes ago
        Don’t forget FaceID and many of the image manipulation.<p>And while everyone else went to more powerful giant LLMs, Apple moved most of Siri from the cloud to your device. Though they do use both (which you can see when Siri corrects itself during transcription—you get the local Siri version corrected later by the cloud version).
    • stetrain2 hours ago
      Apple&#x27;s OSes run a lot of local ML models for many tasks that aren&#x27;t branded as Apple Intelligence, and they have done so for many years now.
    • llm_nerd2 hours ago
      <a href="https:&#x2F;&#x2F;dennisforbes.ca&#x2F;blog&#x2F;microblog&#x2F;2026&#x2F;02&#x2F;apple-neural-engine-and-you&#x2F;" rel="nofollow">https:&#x2F;&#x2F;dennisforbes.ca&#x2F;blog&#x2F;microblog&#x2F;2026&#x2F;02&#x2F;apple-neural-...</a>
    • esafak2 hours ago
      You can convert your own ML models to MLX to use them; Apple Intelligence is not the only application.
      • nullstyle2 hours ago
        MLX does not run on NPUs AFAIK; just gpu and cpu. You have to use CoreML to officially run code on the neural engine.
        • mirsadm2 hours ago
          Even then there is no transparency on how it decides what runs on the ANE&#x2F;GPU etc
          • sroussey28 minutes ago
            Correct. OS level stuff get first priority, so you can’t count on using it.
  • eleventyseven1 hour ago
    &gt; Throughout this series, “we” refers to maderix (human) and Claude Opus 4.6 (by Anthropic) working as a pair. The reverse engineering, benchmarking, and training code were developed collaboratively<p>Sure, &quot;collaboratively.&quot; Why would I ever trust a vibe coded analysis? How do I, a non expert in this niche, know that Opus isn&#x27;t pulling a fast one on both of us? LLMs write convincing bullshit that even fools experts. Have you manually verified each fact in this piece? I doubt it. Thanks for the disclaimer, it saved me from having to read it.
    • Anonbrit33 minutes ago
      Humans also write endless amounts of convincing bullshit, and have done since time immemorial. False papers and faked results have been a growing scourge in academia before LLMs were a thing, and that&#x27;s just counting the intentional fraud - the reproducibility crisis in science, especially medical and psychological science, affects even the best designed and well intentioned of studies.<p>Humans also make mistakes and assumptions while reverse engineering, so it will always need more engineers to go through the results, test things
    • withinboredom53 minutes ago
      Claude likes to hide bad benchmarks from you, so it will show you where you are clearly winning. You even see some weird benchmarks in the article.