iPhone Air has an A19 Pro chip, which is has native matmul full NVIDIA inference speed in laptops, maybe even phones too? news.ycombinator.com/item?id=45186015
Sep 10, 2025 at 10:53 AM
Every Apple M Chip Explained – M1, M2, M3, M4, M5
Apple Design on Twitter / X
iPhone 17 Pro has MORE GRAPHICS POWER than M2 MacBook Air.. this is just BONKERS 🤯 pic.twitter.com/8y6UMNFj82— Apple Design (@TheAppleDesign) September 10, 2025
MacBook Air (M4)
The MacBook Air with the M4 chip is Apple's most powerful yet, with amazing battery life and buttery-smooth performance in a thin and light profile.
MacBook Air (M1, 2020) - Tech Specs - Apple Support
13.3-inch (diagonal) LED-backlit display with IPS technology; 2560-by-1600 native resolution at 227 pixels per inch with support for millions of colors
Gemma 4 QAT models: Optimizing model compression for mobile and laptop efficiency
We’re releasing Gemma 4 quantization-aware training checkpoints, reducing memory requirements and improving on-device performance.
GRID® APPLE A SERIES MOBILE PROCESSORS (V2)
Discover the beauty of Apple's custom-designed chips with our stunning display featuring the A-series chips used in iPhones from 2010 to 2020. Each chip is meticulously arranged and mounted on a sleek acrylic panel, creating a unique and eye-catching piece of tech art. From the groundbreaking A4 chip of the iPhone 4 to the powerful A13 Bionic of the iPhone 11, this display showcases the evolution of Apple's custom silicon over the years. Whether you're a tech enthusiast or a design aficionado, this display is sure to impress. Hang it on your wall, place it on your desk, or gift it to a fellow Apple fan. With its clean design and high-quality materials, this display is a perfect addition to any home, office, or tech collection. Experience the beauty of technology with Grid Studio's A-series chip display. Size: 7.68*4.9*0.4 in (19.5*12.5*1 cm) -Grid Studio
Someone out there now has the chance to run macOS with native A18 Pro driver on his funny iPhone
Someone out there now has the chance to run macOS with native A18 Pro driver on his funny iPhone
Awni Hannun on Twitter / X
According to benchmarks Qwen3.5 4B is as good as GPT 4o.GPT 4o came out ~2 years ago (May 2024).Qwen 3.5 4B runs easily on modern mobile devices.So the gap between frontier intelligence in a datacenter and running a model of equal quality on your iPhone could be 2-3 years.…— Awni Hannun (@awnihannun) March 6, 2026
InferenceMAX™: Open Source Inference Benchmarking
NVIDIA GB200 NVL72, AMD MI355X, Throughput Token per GPU, Latency Tok/s/user, Perf per Dollar, Tokens per Provisioned Megawatt, DeepSeek R1 670B, GPTOSS 120B, Llama3 70B
Argmax - Foundation Models On Device
Run private, real-time, and predictable inference workloads directly on users' devices.
Part 3: iPhone Hardware and How It Powers On-Device AI
By Mirai Labs, frontier on-device AI lab. Building the models, inference runtime, and quantization stack from the device constraint up.
High-Performance Custom Laptops – Upgradeable & Powerful
Fully configurable desktop replacement laptops and mobile workstations. Upgradeable, powerful, portable.
XiongjieDai/GPU-Benchmarks-on-LLM-Inference
Multiple NVIDIA GPUs or Apple Silicon for Large Language Model Inference?
Awni Hannun on Twitter / X
It's very cool that Apple shipped a 20B parameter on-device. You can't put 20B parameters in RAM at any reasonable precision. To make it work they are using pretty exotic architecture by today's standards.A small model predicts from the query (or prompt) which experts to load… pic.twitter.com/Zhe5HcbGuL— Awni Hannun (@awnihannun) June 9, 2026
