Monday, July 27th 2026
TSMC Hits 20,000 Wafers Per Month Milestone on 2 nm Node
TSMC has reportedly reached a new milestone in scaling its 2 nm node high-volume manufacturing, now producing 20,000 wafers per month on the N2 node alone. Manufactured across five fabs in Taiwan, the company currently has only one fab in volume production on the N2 node. This is reportedly Fab 20, which has reached this figure. TSMC began high-volume production of its N2 node in the fourth quarter of 2025. Fast-forward two quarters to today, and the 2 nm node now accounts for only 3% of the company's revenue, while the 3 nm node commands 30%, and the 5 nm node holds the top spot at 33%.
Recently, the company confirmed that the N2 node has four times the number of tape-outs compared to its previous leading-edge 3 nm N3 node. With GAAFET technology, TSMC's N2 delivers up to a 15% performance increase at the same power level or up to a 30% power reduction at the same speed. This has driven four times the number of customers doing tape-outs, including AMD with its EPYC "Venice" server CPUs, Apple with its A20 Pro chips for the iPhone 18 Pro series, and many others. Even though the N2 ramp now accounts for a small portion of revenue and wafer output, the majority of the business is gradually heading in that direction.
For general reference, market-comparable nodes like 18A have similar output. Intel's 18A is produced at two sites: Fab 52 in Phoenix, Arizona, and another in Hillsboro, Oregon. These fabs currently produce about 30,000 wafers per month with the 18A printing. As a reminder, TSMC conducts N2 node manufacturing on ASML's Low-NA EUV machines like the TWINSCAN NXE:3800E, which are expected to increase in price over the next few years. The company has expressed frustration but will comply with the machine purchase.
Source:
UDN
Recently, the company confirmed that the N2 node has four times the number of tape-outs compared to its previous leading-edge 3 nm N3 node. With GAAFET technology, TSMC's N2 delivers up to a 15% performance increase at the same power level or up to a 30% power reduction at the same speed. This has driven four times the number of customers doing tape-outs, including AMD with its EPYC "Venice" server CPUs, Apple with its A20 Pro chips for the iPhone 18 Pro series, and many others. Even though the N2 ramp now accounts for a small portion of revenue and wafer output, the majority of the business is gradually heading in that direction.
4 Comments on TSMC Hits 20,000 Wafers Per Month Milestone on 2 nm Node
Perhaps when/if the can reach 200,000, then maybe the market will settle down a little, and hopefully the prices will go down too, but that's probably a pipe dream given the current economic situation...
Actually, it should have been realized when 2nm generated 3% of revenue. This indicates that N2 is like N3A not a production node, despite claims that it was. The one truly mass-produced is N2P. This clears up the confusion about why N2 had no customers, because in fact, there were none.
By 2028 or 2029 though, AMD and Apple are both rumored to be shifting to A14. It wouldn't be too crazy to assume the other mobile chip companies would be on that node too. So that could free up a good amount of N2 allocation.
Really the supply battle for N2 is prob between Intel and AMD, Apple, Mediatek, Qcomm for their smartphone chips, and maybe some custom AI chips from meta and other hyperscalers.