Apple has announced the A20 Pro, its first iPhone chip built on TSMC’s 2nm N2 process node. The processor represents the biggest manufacturing leap in Apple silicon since the company moved to 3nm with the A17 Pro in 2023, and it arrives at a moment when Samsung has already entered the 2nm race with its own Exynos 2600.
The A20 Pro is expected to ship inside the iPhone 18 Pro, iPhone 18 Pro Max, and, notably, what would be Apple’s first foldable iPhone. All three devices are tied to Apple’s September 9, 2026 “Surprise and Shine” event.
What 2nm actually means for performance
TSMC’s N2 process is projected to deliver a 10-15% speed increase at the same power levels compared to the previous N3E node. Flip that equation around and you get something arguably more useful for a phone: a 25-30% reduction in power consumption while maintaining the same performance.
The A20 Pro also introduces WMCM (Wafer-Level Multi-Chip Module) packaging, which supports a wider 96-bit memory bus. That’s a jump from the 64-bit bus in prior iPhone chips. The result is roughly a 50% increase in memory bandwidth, paired with LPDDR5X memory. The A20 Pro is expected to pack a 7-core GPU. Better thermal management is another anticipated benefit of the new packaging approach.
The competitive landscape at 2nm
Apple isn’t alone in reaching this manufacturing milestone. Samsung released its 2nm Exynos 2600 earlier, making it the first to market with a chip at this node. Apple exclusively uses TSMC, while Samsung manufactures its own Exynos processors in-house.
Qualcomm, which supplies Snapdragon chips to most premium Android phones outside Samsung’s Exynos-equipped models, hasn’t yet announced a 2nm part.
Apple analyst Jeff Pu of Haitong Securities has been tracking the 2nm transition in Apple’s chip roadmap since mid-2025.
Why this matters beyond smartphones
Apple’s A-series processors share DNA with its M-series chips for Mac and iPad. A successful 2nm iPhone chip typically foreshadows a 2nm Mac chip within 12 to 18 months.
For TSMC, Apple’s adoption of the N2 node is commercially critical. Apple is consistently TSMC’s largest customer, and a high-volume product like the iPhone validates the enormous capital expenditure required to bring a new process online.
The 50% memory bandwidth increase is particularly relevant as Apple Intelligence and similar on-device AI systems grow more sophisticated with each iOS update.
Foldable devices face unique thermal and power constraints due to their thinner form factors and flexible displays. A more efficient chip isn’t just nice to have in a foldable, it’s essentially a prerequisite for making the device viable without constant thermal throttling.