NVIDIA GK208

GK208

GK208-301-A1

GK208-320-A1

GK208-400-A1

Block Diagram

SMX Diagram

NVIDIA's GK208 GPU uses the Kepler 2.0 architecture and is made using a 28 nm production process at TSMC. With a die size of 87 mm² and a transistor count of 1,020 million it is a very small chip. GK208 supports DirectX 12 (Feature Level 11_0). For GPU compute applications, OpenCL version 3.0 and CUDA 3.5 can be used. It features 384 shading units, 32 texture mapping units and 8 ROPs.

Graphics Processor

Released: Apr 1st, 2013

GPU Name: GK208

Codename: NV108

Architecture: Kepler 2.0

Foundry: TSMC

Process Size: 28 nm

Transistors: 1,020 million

Density: 11.7M / mm²

Die Size: 87 mm²

Package: FCBGA-595

Graphics Features

DirectX: 12 (11_0)

OpenGL: 4.6

OpenCL: 3.0

Vulkan: 1.2.175

CUDA: 3.5

Shader Model: 6.5 (5.1)

WDDM: 3.0

NVENC: No Support

NVDEC: 1st Gen

PureVideo HD: VP5

VDPAU: Feature Set D

Render Config

Shading Units: 384

TMUs: 32

ROPs: 8

SMX Count: 2

GPCs: 1

Tex L1 Cache: 48 KB per SMX

L1 Cache: 16 KB per SMX

L2 Cache: 512 KB

Max. TDP: 49 W

SFUs: 64

All Kepler 2.0 GPUs

NVIDIA GPU Architecture History

2025 Blackwell 2.0
2024 Blackwell
2023 Hopper
2022-2024 Ada Lovelace
2020-2024 Ampere
2018-2022 Turing
2017-2020 Volta
2016-2021 Pascal
2014-2019 Maxwell 2.0
2014-2017 Maxwell
2013-2015 Kepler 2.0
2012-2018 Kepler
2010-2016 Fermi 2.0
2010-2013 VLIW Vec4
2010-2016 Fermi
2007-2013 Tesla 2.0
2006-2010 Tesla
2003-2013 Curie
2003-2005 Rankine
2001-2003 Kelvin
1999-2005 Celsius
1998-2000 Fahrenheit

Graphics cards using the NVIDIA GK208 GPU

Name	Chip	Memory	Shaders	TMUs	ROPs	GPU Clock	Memory Clock
NVIDIA GeForce GT 720	GK208-320-A1	2 GB	384	32	8	967 MHz	900 MHz
NVIDIA GeForce GT 635 OEM		1024 MB	384	32	8	967 MHz	900 MHz
NVIDIA GeForce GT 740M	GT 740M	2 GB	384	32	8	980 MHz	900 MHz
NVIDIA GeForce GT 735M	N14M-LP	2 GB	384	32	8	575 MHz	900 MHz
NVIDIA GeForce GT 730M	GT 730M	2 GB	384	32	8	719 MHz	900 MHz
NVIDIA GeForce GT 630 Rev. 2	GK208-301-A1	2 GB	384	32	8	902 MHz	900 MHz
NVIDIA GeForce GT 640 Rev. 2	GK208-400-A1	1024 MB	384	32	8	1046 MHz	1252 MHz
NVIDIA GeForce GT 730A	GT 730M	1024 MB	384	32	8	719 MHz	1001 MHz
NVIDIA GeForce GT 630 Rev. 2 PCIe x8	GK208-301-A1	2 GB	384	32	8	902 MHz	900 MHz
NVIDIA Quadro K610M	N15M-Q2-B-A1	1024 MB	192	16	8	954 MHz	650 MHz
NVIDIA Quadro K510M		1024 MB	192	16	8	889 MHz	600 MHz
NVIDIA GeForce 710M		1024 MB	192	16	8	719 MHz	900 MHz
NVIDIA GeForce GT 720M	GT 720M	2 GB	192	16	8	719 MHz	800 MHz
NVIDIA GeForce GT 740A	GT 740M	2 GB	384	32	8	980 MHz	900 MHz
NVIDIA GeForce 825M		1024 MB	384	32	8	850 MHz	900 MHz
NVIDIA GeForce 710A		1024 MB	192	16	8	719 MHz	1001 MHz
NVIDIA GeForce 810A		1024 MB	192	16	8	719 MHz	900 MHz
NVIDIA GeForce 805A		1024 MB	192	16	8	719 MHz	1001 MHz
NVIDIA GeForce GT 710		2 GB	192	16	8	954 MHz	900 MHz

GK208 GPU Notes

NVENC:
NVDEC: 1st Gen
PureVideo HD: VP5
VDPAU: Feature Set D

L1 Cache is configurable from 16 KB up to 48 KB per SMX

Latest Drivers:
Windows XP / Server 2003 x64:
GeForce Release 368.81
Quadro Release R319 U2 (321.01)

Windows Vista:
GeForce Release 365.19
Quadro Release R346 U7 (348.40) / R352 BETA (352.86)

Windows 7 / 8 / 8.1 / 10 / 11 (x32 / x64):
GeForce Release 391.35 / 474.89
Quadro Release R390 U9 (392.37) / R470 U16 (474.82)
Quadro Mobile Release R390 U9 (392.37) / R418 U9 (426.78)

New Forum Posts

22:08 by unclewebb
PL1 keeps throttling even though the wattage is only hitting 35 (6)
22:07 by Solaris17
5700 XT to Arc B580 (10)
22:04 by ShrimpBrime
And so... I bought Arrow Lake (13700k to 265k), my toughts. (11)
21:59 by Hyderz
Extreme temperatures Razer Blade 16 2024 (23)
21:25 by LabRat 891
Motherboard GIGABYTE X870E AORUS ELITE WIFI7 not posting, stuck on code 00 with DRAM red LED turned on (9)
21:07 by bikemanI7
MSI Pro B650 VC WIfi PCIe Lane Question (3)
21:06 by LabRat 891
Help Flash MI50 to Radeon VII Pro (9)
20:52 by unclewebb
Lenovo P16s Gen2 i7-1370P (1)
20:29 by Noci
2 sticks is better then 4 sticks? (66)
20:29 by freeagent
9800x 3d vs 12900k - Battle of the Century (339)