- M5’s 12x neural performance boost marks Apple’s biggest architectural leap
- Dedicated neural accelerators in each GPU core redefine how Apple handles AI processing
- M5 Pro, Max and Ultra should significantly increase neural throughput
Apple’s M-series chips have seen steady performance gains over the past five years, but the jump from M4 to M5 is by far the most notable.
The latest generation chip changes the way Apple handles AI workloads, delivering an increase in neural computing far beyond anything seen before in the tech giant’s silicon.
When Apple’s first chip, the M1, arrived in November 2020, its neural engine could handle around 11 trillion operations per second. The M2 pushed that figure to just under 16, and the M3 jumped to around 18. By the time the M4 arrived last October, that figure had doubled.
Inside Apple Silicon: Part 2 of a 5-part series on M-class processors
This article is the second in a five-part series that takes a deeper dive into Apple’s M-class processors, from the first M1 to the recently announced M5 and our proposed M5 Ultra. Each piece will explore how Apple silicon has evolved in terms of architecture, performance, and design philosophy, and what these changes could mean for the company’s future hardware.
Top of TOPS
With M5, that number skyrocketed to around 133 TOPS, about twelve times the starting point of the M1.
This increase represents the largest increase in the history of Apple’s in-house processors. Rather than relying solely on a faster neural engine, the M5 has a dedicated neural accelerator inside each GPU core.
This allows graphics hardware to support AI workloads directly, distributing inference tasks across the chip instead of passing them through a single engine.
The result is a system that manages model-based processes much more efficiently.
Features like on-device transcription, local image generation, or authoring tools powered by Apple Intelligence all benefit from the new structure.
Every part of the chip now contributes to neural processing, making the overall increase in speed feel less like a step and more like a jump.
On paper, the rest of the chip has also improved. The 10-core processor delivers approximately 15% faster multithreaded performance than the M4, and the unified memory bandwidth reaches 153 GB/s. This supports larger models and more efficient multitasking without increasing power consumption.
The M5 is present in the new 14-inch MacBook Pro and the new iPad Pro. The tablet version uses a nine- or ten-core CPU, depending on storage, but both share the same Neural Engine and GPU layout.
Beyond the chips Apple actually releases, projected numbers for potential future releases suggest how far this design could stretch.
Google Gemini estimates suggest that an M5 Ultra chip could achieve between 600 and 800 TOPS, with the Pro and Max variants falling between 190 and 320.
None of these chips have been announced (and for that matter the M4 Ultra – the M3 Ultra was only announced earlier this year and is in the Mac Studio) and the numbers are just projections, but they follow the growth pattern seen in previous generations, so they’re built on a solid foundation.
Such increases would inevitably raise familiar problems. A desktop-class M5 Ultra would need more cooling and power than Apple’s current compact cases could handle.
What the new M5 shows is that Apple’s chip roadmap is now driven more by neural performance than raw CPU or GPU power. The company has tied its future Macs and iPads to in-device AI. The next generations will decide how far this can evolve before physics and thermals catch up.
Follow TechRadar on Google News And add us as your favorite source to get our news, reviews and expert opinions in your feeds. Make sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp Also.
