- Capability group
- Semiconductors & Advanced Computing
- Domain
- Critical Technology · Hard Power
- Identifier
- factbase.org/cap/EEFHJ9SJ2Q
- Reference
- High-performance computing Wikipedia
Papers
31.6k
2020–2024
Paper growth
+15%
2020–24 on 2015–19
Patent families
33.2k
2020–2024
Filed abroad
37%
families in more than one office
What it covers
What the capability covers and what it leaves to others, and the indicators it will be measured by.
Bounds
Covers computing capability as capacity; the centres as facilities belong to Strategic Infrastructure's data and compute capability, access to frontier compute reads at the deploy and control layers, and the models trained on it sit with Artificial Intelligence.
Lineage
Relocated in v0.1 from the connectivity group (formerly High performance computing) at the same level, carrying its identifier under the relocation convention, and absorbing the accelerator node (retired) from the AI group: hardware for computation is one capacity, and v0.0 held it in two groups, neither of them this chain.
Indicators
- top-ranked supercomputers heldPending
- installed AI accelerator computePending
- domestic accelerator design and productionPending
Who publishes it
Papers classified to the capability, by year and by the nation of their authors' institutions, 2020–2024. Read from the v0.0 research topics High performance computing and AI algorithms and hardware accelerators.
Papers by year
The five leading nations, papers by year
Share of papers, 2020–2024
Of papers with a known author nation; 32% have none and are left out.
- 1United States25.8%6,596 papers
- 2China14.9%3,819 papers
- 3Germany6.0%1,541 papers
- 4India5.1%1,318 papers
- 5United Kingdom4.3%1,105 papers
- 6South Korea3.6%909 papers
- 7Italy3.3%836 papers
- 8Spain3.2%820 papers
- 9Japan2.9%752 papers
- 10France2.7%703 papers
Leading institutions, 2020–2024
By fractional count: each paper divided among its authors' institutions.
- 1National University of Defense Technology246fractional papers
- 2Tsinghua University196fractional papers
- 3Oak Ridge National Laboratory150fractional papers
- 4Peking University129fractional papers
- 5Argonne National Laboratory120fractional papers
- 6Lawrence Berkeley National Laboratory114fractional papers
- 7Chinese Academy of Sciences103fractional papers
- 8University of Science and Technology of China98fractional papers
- 9Lawrence Livermore National Laboratory88fractional papers
- 10Shanghai Jiao Tong University86fractional papers
Who patents it
Patent families by the nation of origin of their applicants and their earliest priority year, from the CPC and IPC classes the capability is read from. 2025 onwards is incomplete: families are published 18 months after priority.
Families by priority year
The five leading origins, families by year
8,539 applicants filed here in 2015–2024; the five largest held 16% of families (HHI 75 of 10,000: unconcentrated).
Share of families by origin, 2020–2024
- 1China53.5%17,762 families
- 2United States21.4%7,086 families
- 3South Korea10.8%3,588 families
- 4Japan3.7%1,236 families
- 5Taiwan2.2%733 families
- 6United Kingdom2.1%713 families
- 7Germany1.4%453 families
- 8France0.8%250 families
- 9Singapore0.6%188 families
- 10Canada0.5%165 families
Applicants filing most internationally, 2015–2024
Families filed in more than one office, by applicant as the offices name them.
- 1INTEL CORP1,615international families
- 2SAMSUNG ELECTRONICS CO LTD1,593international families
- 3HUAWEI TECH CO LTD942international families
- 4QUALCOMM INC567international families
- 5NVIDIA CORP558international families
- 6MICROSOFT TECHNOLOGY LICENSING LLC473international families
- 7ADVANCED RISC MACH LTD376international families
- 8IBM373international families
- 9GOOGLE LLC371international families
- 10FUJITSU LTD351international families
Classification read
- G06N3/06 Hardware for neural networks, including AI accelerator chips · 89.9k documents
- G06F15/76 Processor architectures, including parallel arrays and systems on chip · 76.8k documents
- G06F15/173 Interconnection networks between processors · 28.9k documents
- G06T1/20 Graphics processor architectures · 53.5k documents
- G06F9/3885 Parallel execution units within processors · 43.1k documents
- G06F9/3877 Coprocessors and accelerators · 13.2k documents
Shares inventions with
Capabilities whose classes the same families also fall in, by overlap (Jaccard).
- Machine Learning & Foundation Models9,937 families · 0.083
- Applied AI & Advanced Analytics4,216 families · 0.019
- Semiconductor Design & EDA478 families · 0.016
- Protective Cybersecurity Technologies1,075 families · 0.011
- Quantum Computing258 families · 0.009
Who makes it
The companies that research, design, make, supply and operate the technology, the global leaders and the leading firms of every region.
50 companies in this capability, from 20 nations across 6 regions. Sorted by how experts see each company's global relevance in the field, from very high to very low.
- Very high6 companies
- 1Very highDesignSanta ClaraUnited States199336kNVDA
Dominant designer of AI accelerators (Hopper, Blackwell, Rubin GPUs), NVLink/InfiniBand interconnect and the CUDA software stack that power most AI training clusters and leading supercomputers.
Very high relevance · Design · United States, Santa Clara · founded 1993 · 36k staff
- 2Very highContract manufacturingHsinchuTaiwan198784k2330
Fabricates nearly all leading-edge AI accelerators and HPC processors (NVIDIA, AMD, Google TPU, AWS Trainium) and supplies the CoWoS advanced packaging that limits global AI chip output.
Very high relevance · Contract manufacturing · Taiwan, Hsinchu · founded 1987 · 84k staff
- 3Very highDesignSanta ClaraUnited States196928kAMD
Designs EPYC CPUs and Instinct MI300/MI350 accelerators that power the exascale Frontier and El Capitan systems and a growing share of AI clusters.
Very high relevance · Design · United States, Santa Clara · founded 1969 · 28k staff
- 4Very highComponents & subsystemsIcheonSouth Korea1983–000660
Leading supplier of high-bandwidth memory (HBM3E/HBM4) stacked beside AI accelerators, the main memory constraint on AI compute.
Very high relevance · Components & subsystems · South Korea, Icheon · founded 1983
- 5Very highDesignPalo AltoUnited States196137kAVGO
Co-designs custom AI accelerators (Google TPU and other hyperscaler XPUs) and supplies Tomahawk/Jericho Ethernet switching for AI clusters.
Very high relevance · Design · United States, Palo Alto · founded 1961 · 37k staff
- 6Very highSoftware & platformsMountain ViewUnited States1998183kGOOGL
Designs the Tensor Processing Unit (TPU, now Ironwood) family and operates some of the largest AI training fleets in the world through Google Cloud.
Very high relevance · Software & platforms · United States, Mountain View · founded 1998 · 183k staff
- High13 companies
- 7HighComponents & subsystemsSuwonSouth Korea1969268k005930
Major HBM and DRAM supplier and leading-edge foundry, and among the top patent filers in computing hardware.
High relevance · Components & subsystems · South Korea, Suwon · founded 1969 · 268k staff
- 8HighHardware & devicesShenzhenChina1987208kPrivate
China's leading AI accelerator designer (Ascend 910 series, CloudMatrix 384 supernode) and Kunpeng server CPUs, at the centre of China's domestic AI compute build-out.
High relevance · Hardware & devices · China, Shenzhen · founded 1987 · 208k staff
- 9HighComponents & subsystemsSanta ClaraUnited States1968109kINTC
Supplies Xeon CPUs used across most of the world's HPC installations, Gaudi AI accelerators and the Aurora exascale system, and is the top international patent applicant in this field.
High relevance · Components & subsystems · United States, Santa Clara · founded 1968 · 109k staff
- 10HighSoftware & platformsRedmondUnited States1975228kMSFT
Operates among the largest AI supercomputers through Azure (including OpenAI's training clusters) and designs the Maia AI accelerator and Cobalt CPU.
High relevance · Software & platforms · United States, Redmond · founded 1975 · 228k staff
- 11HighOperations & infrastructureSeattle, WashingtonUnited States19941.6mAMZN
The largest cloud operator; it designs Trainium and Inferentia AI chips (Annapurna Labs) and Graviton CPUs and built Project Rainier for Anthropic.
High relevance · Operations & infrastructure · United States, Seattle, Washington · founded 1994 · 1.6m staff
- 12HighHardware & devicesHoustonUnited States201561kHPE
Through HPE Cray, the leading supercomputer integrator; it built the exascale Frontier, Aurora and El Capitan systems and Europe's LUMI and Alps.
High relevance · Hardware & devices · United States, Houston · founded 2015 · 61k staff
- 13HighComponents & subsystemsBoiseUnited States197848kMU
The only US maker of DRAM and HBM; it supplies HBM3E for NVIDIA and AMD accelerators.
High relevance · Components & subsystems · United States, Boise · founded 1978 · 48k staff
- 14HighContract manufacturingNew TaipeiTaiwan1974–2317
Foxconn (Hon Hai Precision Industry)The largest assembler of AI servers and GB200/GB300 NVL72 racks for NVIDIA and the hyperscalers.
High relevance · Contract manufacturing · Taiwan, New Taipei · founded 1974
- 15HighDesignCambridgeUnited Kingdom19907,096ARM
Its Neoverse CPU architecture underlies NVIDIA Grace, AWS Graviton, Fujitsu A64FX/Monaka and Microsoft Cobalt; it is majority-owned by SoftBank Group, which also owns Graphcore and Ampere.
High relevance · Design · United Kingdom, Cambridge · founded 1990 · 7,096 staff
- 16HighHardware & devicesSan JoseUnited States19935,684SMCI
Major builder of liquid-cooled GPU servers and racks for AI clusters, including xAI's Colossus.
High relevance · Hardware & devices · United States, San Jose · founded 1993 · 5,684 staff
- 17HighHardware & devicesRound RockUnited States1984108kDELL
Large supplier of PowerEdge AI servers and AI-factory systems to CoreWeave, xAI and enterprises, and an HPC system vendor (e.g. TACC's Horizon).
High relevance · Hardware & devices · United States, Round Rock · founded 1984 · 108k staff
- 18HighOperations & infrastructureLivingstonUnited States2017–CRWV
The largest specialist GPU cloud ('neocloud'); it operates hundreds of thousands of NVIDIA GPUs for OpenAI, Microsoft and others.
High relevance · Operations & infrastructure · United States, Livingston · founded 2017
- 19HighSoftware & platformsAustinUnited States1977162kORCL
Oracle Cloud Infrastructure operates zettascale GPU superclusters and is the lead builder of OpenAI's Stargate sites.
High relevance · Software & platforms · United States, Austin · founded 1977 · 162k staff
- Moderate12 companies
- 20ModerateProfessional & data servicesTokyoJapan1935113k6702
Designed the A64FX processor and built the Fugaku supercomputer with RIKEN, and is developing the Monaka Arm CPU for the FugakuNEXT successor.
Moderate relevance · Professional & data services · Japan, Tokyo · founded 1935 · 113k staff
- 21ModerateHardware & devicesBeijingChina198469k0992
Supplies the largest number of TOP500 supercomputers by system count and Neptune liquid-cooled AI servers through its Infrastructure Solutions Group.
Moderate relevance · Hardware & devices · China, Beijing · founded 1984 · 69k staff
- 22ModerateHardware & devicesJinanChina1998–000977
China's largest AI server maker, among the top global server vendors; it is US entity-listed.
Moderate relevance · Hardware & devices · China, Jinan · founded 1998
- 23ModerateDesignBeijingChina2016–688256
Designs the Siyuan (MLU) AI training and inference accelerators, one of China's main domestic alternatives to NVIDIA.
Moderate relevance · Design · China, Beijing · founded 2016
- 24ModerateContract manufacturingShanghaiChina2000–0981
China's most advanced foundry; it makes 7nm-class chips including Huawei Ascend AI accelerators under export controls.
Moderate relevance · Contract manufacturing · China, Shanghai · founded 2000
- 25ModerateSystems integrationBezonsFrance1997–ATO
Europe's main supercomputer builder (BullSequana); it built JUPITER, Europe's first exascale system, with ParTec, and its HPC unit is being acquired by the French state.
Moderate relevance · Systems integration · France, Bezons · founded 1997
- 26ModerateSoftware & platformsArmonkUnited States1911270kIBM
A historic supercomputing leader (Summit, Sierra) that now develops Power processors and the Telum/Spyre AI accelerators, and is a top patent filer in HPC.
Moderate relevance · Software & platforms · United States, Armonk · founded 1911 · 270k staff
- 27ModerateOperations & infrastructureHangzhouChina1999124kBABA
China's largest cloud provider (Alibaba Cloud); it designs AI inference chips and Yitian CPUs through its T-Head unit.
Moderate relevance · Operations & infrastructure · China, Hangzhou · founded 1999 · 124k staff
- 28ModerateHardware & devicesSunnyvaleUnited States2016––
Builds the wafer-scale WSE-3 AI processor and CS-3 systems, including the Condor Galaxy supercomputers built with G42.
Moderate relevance · Hardware & devices · United States, Sunnyvale · founded 2016
- 29ModerateDesignTorontoCanada2016––
Founded in Toronto, it designs RISC-V-based AI accelerators (Wormhole, Blackhole) and licenses CPU and AI IP, led by Jim Keller.
Moderate relevance · Design · Canada, Toronto · founded 2016
- 30ModerateOperations & infrastructureAmsterdamNetherlands2024–NBIS
European-headquartered AI GPU cloud operator (the former Yandex N.V.); it runs large NVIDIA clusters in Finland, the US and elsewhere and has a multibillion-dollar Microsoft contract.
Moderate relevance · Operations & infrastructure · Netherlands, Amsterdam · founded 2024
- 31ModerateOperations & infrastructureAbu DhabiUnited Arab Emirates2018–Private
The UAE's AI champion; it operates the Condor Galaxy AI supercomputers with Cerebras through Core42 and leads the 5GW UAE–US AI campus and Stargate UAE.
Moderate relevance · Operations & infrastructure · United Arab Emirates, Abu Dhabi · founded 2018
- Low12 companies
- 32LowDesignTianjinChina2014–688041
Designs x86-compatible Hygon CPUs (licensed from AMD's Zen) and DCU GPGPU accelerators used in Chinese supercomputers and AI data centres.
Low relevance · Design · China, Tianjin · founded 2014
- 33LowHardware & devicesTianjinChina2006–603019
Sugon (Dawning Information Industry)China's leading supercomputer builder (Dawning series and national computing-power networks), backed by the Chinese Academy of Sciences.
Low relevance · Hardware & devices · China, Tianjin · founded 2006
- 34LowDesignBeijingChina2020–688795
Chinese GPU designer (MTT S4000 and KUAE AI clusters) that listed on the STAR Market in December 2025.
Low relevance · Design · China, Beijing · founded 2020
- 35LowProfessional & data servicesTokyoJapan1899105k6701
Long-standing vector supercomputer maker (Earth Simulator, SX-Aurora TSUBASA) and HPC system integrator for Japanese and European centres.
Low relevance · Professional & data services · Japan, Tokyo · founded 1899 · 105k staff
- 36LowDesignSeongnamSouth Korea2020––
South Korea's leading AI inference chip startup (ATOM, REBEL); it merged with SK Group's Sapeon in 2024.
Low relevance · Design · South Korea, Seongnam · founded 2020
- 37LowSystems integrationMunichGermany1999–JY0
Developer of modular supercomputing architecture and co-builder, with Eviden, of the JUPITER exascale system at Jülich.
Low relevance · Systems integration · Germany, Munich · founded 1999
- 38LowDesignMaisons-LaffitteFrance2019––
Designs Rhea, the European Processor Initiative's Arm-based HPC CPU intended for European exascale systems.
Low relevance · Design · France, Maisons-Laffitte · founded 2019
- 39LowDesignGivatayimIsrael2017––
Israeli developer of the Maverick dataflow accelerator for HPC, deployed at Sandia National Laboratories.
Low relevance · Design · Israel, Givatayim · founded 2017
- 40LowOperations & infrastructureRiyadhSaudi Arabia2025–State-owned
PIF-owned Saudi AI company building national AI compute, with deals for hundreds of thousands of NVIDIA GPUs plus AMD and AWS capacity.
Low relevance · Operations & infrastructure · Saudi Arabia, Riyadh · founded 2025
- 41LowHardware & devicesFaridabadIndia1999–NETWEB
India's leading domestic HPC and AI server maker (Tyrone systems); it builds GPU servers locally under the IndiaAI mission and supplies Indian supercomputers.
Low relevance · Hardware & devices · India, Faridabad · founded 1999
- 42LowResearch & IPPuneIndia1988–State-owned
C-DAC (Centre for Development of Advanced Computing)Indian government R&D body behind the PARAM supercomputers and the National Supercomputing Mission, including the AIRAWAT AI system and indigenous Rudra servers.
Low relevance · Research & IP · India, Pune · founded 1988
- 43LowOperations & infrastructureMumbaiIndia2019–Private
Hiranandani Group's GPU cloud (Shakti Cloud), India's largest NVIDIA H100 deployment and a main supplier of compute to the IndiaAI mission.
Low relevance · Operations & infrastructure · India, Mumbai · founded 2019
- Very low7 companies
- 44Very lowOperations & infrastructureKuala LumpurMalaysia1993–6742
Through YTL AI Cloud, it operates NVIDIA Grace Blackwell AI supercomputing in Johor, among the largest AI compute deployments in Southeast Asia.
Very low relevance · Operations & infrastructure · Malaysia, Kuala Lumpur · founded 1993
- 45Very lowProfessional & data servicesHanoiVietnam1988–FPT
FPT Corporation (FPT Semiconductor)Vietnam's largest IT group; it runs the FPT AI Factory GPU clouds (NVIDIA H100/H200) in Vietnam and Japan.
Very low relevance · Professional & data services · Vietnam, Hanoi · founded 1988
- 46Very lowOperations & infrastructureSingaporeSingapore1879–Z74
Runs RE:AI, a regional GPU-as-a-service cloud built with NVIDIA, through its Nxera data centre arm.
Very low relevance · Operations & infrastructure · Singapore, Singapore · founded 1879
- 47Very lowOperations & infrastructureJakartaIndonesia1967–ISAT
Built Indonesia's sovereign AI GPU cloud with NVIDIA (Lintasarta GPU Merdeka) and is the country's main AI compute provider.
Very low relevance · Operations & infrastructure · Indonesia, Jakarta · founded 1967
- 48Very lowOperations & infrastructureBangkokThailand2023––
Thailand's first NVIDIA Cloud Partner; it operates H100/H200 GPU clusters for a Thai sovereign AI cloud.
Very low relevance · Operations & infrastructure · Thailand, Bangkok · founded 2023
- 49Very lowChemicals & gasesRio de JaneiroBrazil1953–PETR4
Operates Latin America's largest supercomputers (Pegaso, Harpia and successors on the TOP500) for seismic processing.
Very low relevance · Chemicals & gases · Brazil, Rio de Janeiro · founded 1953
- 50Very lowOperations & infrastructureLondonUnited Kingdom2021–Private
Strive Masiyiwa's pan-African technology group (Econet lineage); it is deploying Africa's first NVIDIA 'AI Factory' in its South African data centres, with expansion planned to Nigeria, Egypt, Kenya and Morocco.
Very low relevance · Operations & infrastructure · United Kingdom, London · founded 2021
