According to industry sources, TSMC's next-generation 2nm (nanometer) process node is entering mass production, with monthly capacity projected to reach 200,000 wafers by the end of 2028.
Production will be spearheaded by TSMC's Baoshan Fab 20 in Hsinchu, expected to achieve a monthly output of 40,000 to 45,000 wafers by Q4 2025 and scale to 55,000–60,000 wafers by late 2026 to 2027. Expansion will continue at Fab 22 in Kaohsiung, which will become the key site for 2nm capacity. Built in six phases, the P2 plant alone is set to reach 10,000–15,000 wafers per month by late 2025, followed by phased deployments of P3 through P6, pushing monthly output to approximately 50,000–55,000 wafers in 2026, 80,000 wafers in 2027, and up to 145,000–150,000 wafers by the end of 2028.
Collectively, TSMC's 2nm monthly capacity is projected to exceed 100,000 wafers by the end of 2026 and nearly double to 200,000 wafers by 2028. TSMC also plans to manufacture 2nm chips at its future facilities in the United States.
Major global semiconductor players—such as AMD, Apple, Qualcomm, MediaTek, Marvell, Broadcom, and Bitmain—are expected to be key adopters of TSMC's 2nm technology.
Industry analysts highlight that while TSMC traditionally begins each new node with an initial monthly output around 50,000 wafers and gradually scales to 140,000–150,000, the aggressive jump to 200,000 wafers for 2nm demonstrates a bold, strategic investment.
This leap is driven by the adoption of the advanced GAA (Gate-All-Around) transistor architecture, which involves significantly higher capital expenditure. While competitors like Samsung, Intel, and Rapidus are still grappling with yield, scalability, and production stability at similar nodes, TSMC's early investment gives it a solid lead in the 2nm race.