China has returned to the top of the global supercomputer rankings for the first time since 2017, with LineShine, a new system installed at the National Supercomputing Centre in Shenzhen, displacing the United States’ El Capitan as the world’s most powerful computer on the authoritative TOP500 list. Announced at the ISC 2026 conference in Hamburg, Germany, on June 23, LineShine achieved 2.198 exaflops on the High Performance Linpack benchmark, approximately 80 percent of its 2.736 exaflop theoretical peak, making it the first system on the TOP500 to exceed two exaflops of sustained double-precision performance using a CPU-only architecture. The system was built by the Shenzhen Cloud Computing Center and is based on a custom Chinese processor and the LingKun platform.
LineShine’s design drew particular attention because it does not rely on graphics processing units, which have become central to artificial intelligence workloads and advanced computing and have been the primary focus of US export restrictions aimed at limiting China’s access to the most powerful semiconductor technology. Instead, LineShine uses a CPU-only architecture built on Chinese-developed technology, running on China’s LingKun platform, LingQi interconnect network, and Kylin operating system. The system achieved a 20 percent performance lead over El Capitan, which had held the top position since November 2024 and is housed at Lawrence Livermore National Laboratory, where the US government uses it to develop and maintain its nuclear weapons stockpile.
The achievement is technically impressive but comes with important caveats that experts have been quick to highlight. Technology and policy experts said the results do not mean that China has the world’s fastest computer for artificial intelligence work, because the TOP500 ranking uses the High Performance Linpack benchmark, which measures one specific kind of mathematical calculation that does not reflect AI training workloads. LineShine ranked fourth on a separate benchmark test designed to simulate computing work more similar to artificial intelligence, while El Capitan remained the number one system on the HPL-MxP mixed-precision benchmark at 16.7 exaflops, a measure that better reflects AI-adjacent workloads. The LineShine system carries no advanced AI chips, likely because the tools needed to manufacture those chips remain under US export controls.
China first took the top spot on the TOP500 in 2010 and traded the title back and forth with the US and Japan until 2023, when China stopped submitting its systems amid years of chip and computing-related export controls that restricted its access to the most advanced semiconductors. The three-year absence from the rankings, and the decision to re-enter now with an entirely domestically produced system, reflects China’s broader push to demonstrate technological self-sufficiency in advanced computing. Jack Dongarra, an organiser of the TOP500 rankings, said the system showed China capable of holding its own in advanced computing despite US export restrictions, while Addison Snell of Intersect360 Research noted that the US still leads globally in technology but that the gap is not wide and with the rapid pace of evolution, the global order could change quickly. For the ongoing US-China technology competition, LineShine’s return to the top of the rankings is as much a geopolitical statement as a scientific achievement, demonstrating that US export controls, while effective at limiting China’s access to GPU-based AI computing, have not prevented Beijing from building world-record high-performance computing infrastructure using its own chips and interconnects.
Follow the SPIN IDG WhatsApp Channel for updates across the Smart Pakistan Insights Network covering all of Pakistan’s technology ecosystem.
