Report an Error

NVIDIA GV10B

Default
GV10B
Die Shot
Die Shot
SM Diagram
SM Diagram
NVIDIA's GV10B GPU uses the Volta architecture and is made using a 12 nm production process at TSMC. With a die size of 350 mm² and a transistor count of 9,000 million it is a large chip. GV10B supports DirectX 12 (Feature Level 12_1). For GPU compute applications, OpenCL version 1.2 and CUDA 7.2 can be used. It features 512 shading units, 32 texture mapping units and 16 ROPs. Also included are 64 tensor cores which help improve the speed of machine learning applications.

Graphics Processor

Released
Oct 2018
GPU Name
GV10B
Architecture
Volta
Foundry
TSMC
Process Size
12 nm
Transistors
9,000 million
Density
25.7M / mm²
Die Size
350 mm²

Graphics Features

DirectX
12 (12_1)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.2
CUDA
7.2
Shader Model
6.0
Tensor Cores
1st Gen
PureVideo HD
VP9
VDPAU
Feature Set I

Render Config

Shading Units
512
TMUs
32
ROPs
16
SM Count
8
Tensor Cores
64
SFUs
128
TPCs
4
GPCs
1
L1 Cache
128 KB per SM
L2 Cache
512 KB
Max. TDP
30 W

All Volta GPUs

NVIDIA GPU Architecture History

Graphics cards using the NVIDIA GV10B GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
NVIDIA Jetson AGX Xavier 16 GB
16 GB 512 32 16 854 MHz 1377 MHz 2133 MHz
NVIDIA Jetson Xavier NX 8 GB
8 GB 384 24 16 854 MHz 1100 MHz 1866 MHz
NVIDIA Jetson Xavier NX 16 GB
16 GB 384 24 16 854 MHz 1100 MHz 1866 MHz
NVIDIA Jetson AGX Xavier 32 GB
32 GB 512 32 16 854 MHz 1377 MHz 2133 MHz

GV10B GPU Notes

Tensor Cores: 1st Gen
PureVideo HD: VP9
VDPAU: Feature Set I
May 23rd, 2024 19:32 EDT change timezone

Latest GPU Drivers

New Forum Posts

Popular Reviews

Controversial News Posts