Leading FinFET: 14 to 6 nm
Zen 2 and 3, the A12 to A14, the A100 and the EUV lithography that came with 7 nm. 90 chips.
KindCPUGPUAI acceleratorSoCMicrocontrollerMemoryFPGADSPNetwork chipModemSensorAnalogConsole chipMining or securityQuantum
🇺🇸 United StatesVolta
Nvidia's 2017 GPU architecture, first with Tensor Cores for deep learning.GPU · GPU architecture · 2017
🇺🇸 United StatesPOWER9
IBM's Power ISA server microprocessor family, built on 14 nm FinFET siliconCPU · Power · 2016
🇯🇵 JapanFujitsu A64FX
48-core ARM supercomputer CPU with 512-bit SVE and HBM2, built for FugakuCPU · Arm · 2019
WorldEpyc
AMD's server and embedded x86-64 CPUs built on Zen coresCPU · x86 · 2017
🇺🇸 United StatesXeon Phi
Intel's x86 manycore coprocessor family for supercomputers and HPCCPU · x86 · 2010
🇺🇸 United StatesSapphire Rapids
Intel server and workstation chips with chiplet designCPU · x86 · 2023
🇺🇸 United StatesCascade Lake
Intel 14 nm server and workstation processor generationCPU · x86 · 2019
WorldGDDR7 SDRAM
High-bandwidth graphics memory standard for GPUs and computing.Memory · 2024
WorldGDDR6 SDRAM
High-bandwidth graphics memory for GPUs and consolesMemory · 2017
🇺🇸 United StatesCDNA
AMD microarchitecture for data centre GPU acceleratorsGPU · GPU architecture · 2020
🇺🇸 United StatesPower10
IBM's 7 nm server CPU family for AI-heavy enterprise workloadsCPU · Power · 2021
🇺🇸 United States and 1 moreAMD Instinct MI100
CDNA architecture GPU for high-performance computing and AI workloads.GPU · GPU architecture · 2020
🇺🇸 United StatesTesla V100
Nvidia Volta GPU for AI and HPC data centresGPU · GPU architecture · 2017
🇺🇸 United StatesAMD Instinct MI250X
Dual-GPU accelerator for supercomputersGPU · GPU architecture
🇺🇸 United StatesIBM Power
RISC microprocessors for servers and supercomputers by IBMCPU · Power · 1990
Photo credits on this page
The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.
- Power10: Henriok, CC0, source
- POWER9: IBM España, Public domain, source
- Epyc: Smial ( talk ), FAL, source
- Xeon Phi: Capitalistcakes, CC0, source
- Sapphire Rapids: 极客湾Geekerwan, CC BY 3.0, source
- Cascade Lake: Smial ( talk ), FAL, source
- GeForce RTX 30 series: PantheraLeo1359531, CC BY 4.0, source
- GDDR7 SDRAM: 极客湾Geekerwan, CC BY 3.0, source
- IBM Power: Henriok, CC0, source