Inside the M4 Apple Neural Engine, Part 1: Reverse Engineering
How we bypassed CoreML and talked directly to the hardware
A note on “we”:
Throughout this series, “we” refers to maderix (human) and Cla...
Similar Articles (10 found)
🔍 65.4% similar
Every year, we have a new iPhone that claims to be faster and better in every way. And yes, these new computer vision models and new image sensors can...
🔍 63.2% similar
Understanding LLM Inference Engines: Inside Nano-vLLM (Part 1)
Architecture, Scheduling, and the Path from Prompt to Token
When deploying large langua...
🔍 61.4% similar
If you use an Apple silicon Mac I’m sure you have been impressed by its performance. Whether you’re working with images, audio, video or building soft...
🔍 60.8% similar
By John Gruber
npx workos: An AI agent that writes auth directly into your codebase.
Apple Newsroom:
Apple today announced M5 Pro and M5 Max, the worl...
🔍 59.1% similar
Techniques for training large neural networks
Large neural networks are at the core of many recent advances in AI, but training them is a difficult en...
🔍 58.2% similar
Claude Opus 4.5, and why evaluating new LLMs is increasingly difficult
24th November 2025
Anthropic released Claude Opus 4.5 this morning, which they ...
🔍 57.5% similar
How We Cut Inference Costs from $46K to $7.5K Fine-Tuning Qwen-Image-Edit
Running quality inference at scale is something we think about a lot at Oxen...
🔍 57.1% similar
v1.0 - Initial draft
What I’ll talk about #
I started Didact AI in 2019, out of a firm belief that stock picking engines built on top of machine learn...
🔍 56.8% similar
The 2018 Mac Mini
I hardly ever think about my Mac Mini, but it serves a vital role for my family as our home-theater mixer, Plex server, ScanSnap ser...
🔍 56.4% similar
💜Qwen3.5 - How to Run Locally Guide
Run the new Qwen3.5 LLMs including Medium: Qwen3.5-35B-A3B, 27B, 122B-A10B, Small: Qwen3.5-0.8B, 2B, 4B, 9B and 39...