Skip to content
IFAI

Industry

Apple drops M4 to turn iPad Pro into an AI beast

Published Reach 41

Apple just unveiled its next-generation M4 chip, debuting inside the redesigned iPad Pro. Built on second-generation 3nm technology, the processor features Apple's fastest Neural Engine yet, claiming to outpace the NPU of any AI PC today. Along with hardware-accelerated ray tracing and dynamic caching, Apple is positioning the tablet as an on-device AI powerhouse.

Apple drops M4 to turn iPad Pro into an AI beast
Image: apple.com

Apple just pulled an unexpected move. Instead of bringing its next-generation silicon to the Mac first, the company unveiled the M4 chip inside an impossibly thin new iPad Pro.

Built on second-generation 3-nanometer technology, the M4 packs 28 billion transistors and introduces major upgrades across the CPU, GPU, display engine, and Neural Engine.

More power, way less power draw

The CPU moves up to 10 cores, split into 4 performance cores and 6 efficiency cores. Apple says this setup delivers up to 1.5x faster CPU performance compared to the M2 in the previous iPad Pro.

Efficiency is the headline here. M4 can match M2 performance using just half the power. Put next to a thin-and-light laptop PC chip like the Intel Core Ultra 7 155H, M4 delivers identical performance using only a fourth of the power.

The 10-core GPU borrows directly from the M3 architecture. It brings Dynamic Caching to the iPad for the first time, alongside hardware-accelerated ray tracing and mesh shading. Professional 3D rendering in apps like Octane is up to 4x faster than on M2.

There is also a brand-new display engine built specifically to run the tandem OLED Ultra Retina XDR display, plus hardware acceleration for AV1 video decoding.

Aiming straight at the AI PC crowd

Apple spent serious time highlighting the Neural Engine, which hits 38 trillion operations per second (TOPS). That makes it 60 times faster than the original Neural Engine found in the A11 Bionic.

Apple explicitly claimed that this NPU outperforms the neural processing unit found in any current AI PC on the market.

In real workflows, that means isolating a subject from the background across a 4K clip in Final Cut Pro with a single tap, or generating sheet music in real time inside StaffPad while listening to a piano performance.

Why it matters

Skipping the Mac and debuting M4 inside an iPad Pro tells you everything you need to know about where the industry is heading. With Qualcomm and Intel pushing the AI PC narrative hard, Apple needed to plant a flag and assert hardware dominance before Copilot+ PCs flooded the market.

For power users doing heavy video editing or 3D work on the go, the performance jump is real and the thermals look remarkably tight. But for everyday users, the hardware is honestly overkill.

We think the smart part was using a sub-6mm tablet to prove power efficiency. The issue, however, remains the software. Putting a desktop-class chip with a 38 TOPS NPU into an iPad creates a wild mismatch until iPadOS genuinely allows pros to take the training wheels off.

Sources (1)

  1. [1]
    apple.com(opens in a new tab)Originalapple.com

Best of AI, daily