News
EveryChip

GPU architectures

Nvidia Tesla to Blackwell, AMD GCN, RDNA and CDNA, Intel Xe, Arm Mali and Imagination PowerVR. 134 chips on EveryChip so far.

GPU architectures are the design generations of graphics and compute processors. Nvidia names each one after a scientist (Tesla, Fermi, Kepler, Pascal, Volta, Turing, Ampere, Hopper, Blackwell, Rubin), AMD uses families such as Graphics Core Next, RDNA and CDNA, and Intel has Xe. Each generation changes how the many small cores are grouped, what special units they carry and how they reach memory.

Generation by generation

Nvidia's Tesla (2006) unified the shader units and brought CUDA. Volta (2017) added tensor cores for matrix maths, Turing (2018) added ray tracing cores for realistic lighting, and Hopper (2022) and Blackwell (2024) were built for AI. AMD's Graphics Core Next (2012) served both games and compute; in 2019 and 2020 AMD split its line into RDNA for games and CDNA for data centres.

Programming models

GPU instruction sets change with each generation, so programs are written in higher-level languages such as CUDA, OpenCL, HIP or shader languages, and a driver compiles them for the chip. Nvidia's long investment in CUDA software is one reason for its lead in AI. Mobile GPUs from Arm (Mali), Imagination (PowerVR), Qualcomm (Adreno) and Apple use tile-based designs that save memory traffic.

Famous GPU architecture chips

Further reading

WorldGeForce 900 series Nvidia graphics cards using the Maxwell architecture on a 28 nm process.GPU · GPU architecture · 2014 WorldGraphics Core Next AMD graphics processor architecture introduced in 2012GPU · GPU architecture · 2012 Radeon 🇺🇸 United StatesRadeon Brand of graphics processors and computer products by AMDGPU · GPU architecture · 2000 🇺🇸 United StatesRDNA AMD graphics microarchitecture for GPUs and consolesGPU · GPU architecture · 2019 TeraScale 🇺🇸 United StatesTeraScale ATI and AMD graphics microarchitecture using VLIW designGPU · GPU architecture · 2007 🇨🇦 CanadaRadeon R100 First generation Radeon graphics chip with hardware T&LGPU · GPU architecture · 2000 🇨🇦 CanadaRadeon R200 Second-generation ATI GPU with DirectX 8.1 supportGPU · GPU architecture · 2001 🇨🇦 CanadaRadeon R300 First consumer GPU with full Direct3D 9 supportGPU · GPU architecture · 2002 WorldRadeon HD 2000 series ATI's first DirectX 10 graphics card family with unified shaders.GPU · GPU architecture · 2007 🇨🇦 CanadaRadeon X1000 series ATI graphics cards from 2005 using the R500 architectureGPU · GPU architecture · 2005 🇨🇦 CanadaRadeon HD 3000 series ATI graphics cards using TeraScale 1 architecture and 55 nm process.GPU · GPU architecture · 2007 🇺🇸 United StatesRadeon HD 6000 series AMD graphics processor family using TeraScale architectureGPU · GPU architecture · 2010 Radeon HD 8000 series WorldRadeon HD 8000 series OEM rebadged graphics processors from AMDGPU · GPU architecture 🇺🇸 United StatesRadeon 200 series AMD graphics processors using GCN and TeraScale architectures.GPU · GPU architecture · 2013 🇺🇸 United StatesRadeon HD 5000 series First DirectX 11 graphics processor family from AMDGPU · GPU architecture · 2009 🇺🇸 United StatesRadeon 300 series AMD graphics processors using GCN architecture and 28 nm processGPU · GPU architecture · 2015 🇺🇸 United StatesRadeon HD 7000 series AMD graphics processor family using 28 nm process and GCN architectureGPU · GPU architecture · 2012 Radeon RX 400 series 🇺🇸 United StatesRadeon RX 400 series AMD graphics processors using Polaris chips and GCN 4 architectureGPU · GPU architecture · 2016 Radeon RX Vega 🇺🇸 United StatesRadeon RX Vega AMD graphics processors using the fifth generation GCN architectureGPU · GPU architecture · 2017 Radeon RX 6000 series 🇺🇸 United StatesRadeon RX 6000 series AMD RDNA 2 graphics cards with ray tracingGPU · GPU architecture · 2020 Radeon RX 5000 series WorldRadeon RX 5000 series AMD graphics processors using RDNA architecture and 7 nm processGPU · GPU architecture · 2019 🇺🇸 United StatesRadeon RX 9000 series Mainstream consumer graphics cards using RDNA 4 architectureGPU · GPU architecture · 2025 🇺🇸 United StatesRadeon RX 7000 series AMD RDNA 3 graphics cards with chiplet designGPU · GPU architecture · 2022 AMD FirePro 🇺🇸 United StatesAMD FirePro Professional graphics cards for workstations and servers by AMDGPU · GPU architecture 🇺🇸 United StatesIntel Arc Discrete graphics processors by Intel for gaming and workstationsGPU · GPU architecture · 2022 WorldGeForce 256 Nvidia's first GeForce, a 1999 3D accelerator with hardware transform and lightingGPU · GPU architecture · 1999 Intel Xe 🇺🇸 United StatesIntel Xe Graphics architecture for integrated and discrete GPUsGPU · GPU architecture · 2020 WorldMatrox Parhelia A 2002 GPU known for triple-monitor support and high image quality.GPU · GPU architecture · 2002 🇺🇸 United StatesVoodoo2 3D graphics accelerator chipset with dual-texturing and SLI supportGPU · GPU architecture · 1998 WorldMatrox G400 Dual-monitor graphics card with high-quality 2D outputGPU · GPU architecture · 1999 🇺🇸 United StatesVoodoo3 Gaming graphics card series by 3dfx InteractiveGPU · GPU architecture · 1999 🇺🇸 United StatesVoodoo 5 Final graphics card series from 3dfx InteractiveGPU · GPU architecture · 2000 WorldS3 ViRGE Early mass-market 2D and 3D graphics accelerator chip from S3.GPU · GPU architecture · 1996 🇺🇸 United StatesS3 Savage PC graphics chipset series by S3 GraphicsGPU · GPU architecture · 1998 AMD Radeon 600 series 🇺🇸 United StatesAMD Radeon 600 series Entry-level OEM graphics cards based on older Polaris architecture.GPU · GPU architecture · 2019 🇺🇸 United StatesGeForce RTX 20 series First consumer GPUs with hardware ray tracingGPU · GPU architecture · 2018 Ampere 🇺🇸 United StatesAmpere Nvidia's 2020 GPU architecture behind GeForce RTX 30 and the A100 acceleratorGPU · GPU architecture · 2020 NVENC WorldNVENC Dedicated video encoding hardware in Nvidia GPUsGPU · GPU architecture · 2012 WorldSiS 6326 Low-cost GPU for corporate motherboards and basic 3D tasks.GPU · GPU architecture · 1997 🇺🇸 United StatesRIVA TNT Dual-pipeline PC graphics chip from NvidiaGPU · GPU architecture · 1998 WorldGeForce 100 series OEM rebranded Tesla graphics cards from 2009GPU · GPU architecture · 2009 🇺🇸 United StatesGeForce 800M series Nvidia laptop graphics chips from 2014 using Fermi, Kepler and Maxwell designs.GPU · GPU architecture · 2014 Quadro Plex WorldQuadro Plex External multi-GPU box for professional 3D workGPU · GPU architecture

Photo credits on this page

The small pictures above come from Wikimedia Commons under free licences. Some are cropped to fit.

The weekly digest

Every Monday: the chip of the day, new pages on EveryChip, and the conferences and anniversaries of the week ahead.