LineShine, a previously unlisted system installed in China, debuts at No. 1 in the 67th edition of the TOP500 list of the world’s most powerful supercomputers.
The company displaced El Capitan as the world’s most powerful supercomputer as measured by the High Performance Linpack (HPL) benchmark.
The new list also reflects continued depth in United States and European exascale capability, a new entrant in Italy’s HPC fleet, and unchanged leadership atop the Green500 energy-efficiency ranking.
LineShine achieved 2.198 Exaflop/s on HPL — about 80% of its 2.736 Exaflop/s theoretical peak — making it the first system on the TOP500 to exceed two exaflops of sustained double-precision performance using CPUs only.
Installed at the National Supercomputing Centre in Shenzhen (NSCS) and built by the Shenzhen Cloud Computing Center, the system is based on a custom Chinese processor and the “LingKun” platform: 13.79 million cores across 304-core LX2 processors running at 1.55 GHz, linked by the proprietary LingQi interconnect and running Kylin OS.
LineShine draws approximately 42.2 megawatts of power, for an efficiency of 52.07 Gigaflops/Watt. Its debut marks the first time since 2017 that a Chinese system has led the TOP500, and it also takes over the No. 1 position on the HPCG ranking with 22.00 HPCG-Petaflop/s.
On the HPL-MxP mixed-precision benchmark, LineShine reached 7.92 Exaflop/s for fourth place, a comparatively modest 3.6x speedup over its HPL score that points to a CPU-only design without dedicated low-precision accelerators.
LineShine’s debut increases the number of systems sustaining more than one exaflop/s on HPL from four to five and, for the first time, places exascale systems across Asia, North America, and Europe simultaneously.
El Capitan, at Lawrence Livermore National Laboratory, drops to No. 2 but is otherwise unchanged at 1.809 Exaflop/s, 11.34 million cores, and 60.94 Gigaflops/Watt, built on the HPE Cray EX255a architecture with AMD 4th Gen EPYC CPUs and AMD Instinct MI300A accelerators.
Frontier, at Oak Ridge National Laboratory, moves to No. 3 at 1.353 Exaflop/s, and Aurora, at Argonne National Laboratory, holds No. 4 at 1.012 Exaflop/s. JUPITER Booster, operated by the Jülich Supercomputing Centre under the EuroHPC Joint Undertaking, moves to No. 5 at exactly 1.000 Exaflop/s, remaining Europe’s only system above the exascale threshold on HPL.
Eni SpA’s new HPC7 system enters the list directly at No. 6 with 571.5 Petaflop/s, built on the same HPE Cray EX255a / AMD Instinct MI300A architecture as El Capitan, and becomes the most powerful machine in Eni’s HPC fleet alongside its existing HPC6 system.
Microsoft’s Azure-based Eagle system falls to No. 7 at 561.2 Petaflop/s, followed by HPC6 at No. 8 (477.9 Petaflop/s). Japan’s Fugaku holds No. 9 at 442 Petaflop/s, and Switzerland’s Alps system rounds out the Top 10 at No. 10 with 434.9 Petaflop/s.
Finland’s LUMI and Italy’s Leonardo, No. 9 and No. 10 last edition, fall just outside the new Top 10 at No. 11 and No. 12, respectively.
Architectural and vendor diversity in the Top 10
The June 2026 Top 10 illustrates an unusually high degree of architectural diversity, reflecting the increasingly heterogeneous nature of high-performance computing.
The systems span custom Chinese architectures (LineShine’s LingKun processors and LingQi interconnect), AMD-based systems ranging from exascale (El Capitan and Frontier) to sub-exaflop performance (HPC7 and HPC6), an Intel-based exascale design (Aurora), NVIDIA Grace Hopper architecture (JUPITER Booster and Alps), Microsoft’s cloud-based Eagle system combining Intel Xeon processors with NVIDIA H100 accelerators, and Japan’s distinctive Fugaku system built around Fujitsu’s A64FX Arm processors.
The list demonstrates that there is no single dominant technology path to leadership-class computing; instead, vendors are pursuing a variety of CPU, GPU, APU, and custom-accelerator approaches coupled with different interconnect and system designs.
Looking at vendor representation, HPE is the dominant system integrator, supplying six of the ten systems (El Capitan, Frontier, Aurora, HPC7, HPC6, and Alps); Aurora runs on the HPE Cray EX platform but is credited to Intel as prime contractor.
On the processor side, AMD has the strongest presence, powering four systems directly (El Capitan, Frontier, HPC7, and HPC6) and contributing more than 40 percent of the combined Top 10 HPL performance. NVIDIA technology appears in three systems (JUPITER Booster, Eagle, and Alps), while Intel is represented both as a complete platform vendor (Aurora) and through Xeon processors in Eagle.
Overall, the Top 10 reflects a competitive landscape led by HPE integration expertise, AMD’s strong position in exascale computing, NVIDIA’s growing influence through AI-oriented accelerators, and continued innovation from national computing programs in China, Japan, Europe, and the United States.
On the HPCG benchmark, which measures performance on data-intensive, real-world application patterns rather than raw floating-point throughput, LineShine takes over the No. 1 position with 22.00 HPCG-Petaflop/s, ahead of El Capitan (17.41) and Fugaku, now third (16.00). Frontier holds fourth (14.05), Eni’s new HPC7 system takes fifth (5.95), and Aurora rounds out the top six (5.61). JUPITER Booster has not yet submitted an HPCG result.
