Report an Error

NVIDIA GA100

Default
GA100
Block Diagram
Block Diagram
SM Diagram
SM Diagram
NVIDIA's GA100 GPU uses the Ampere architecture and is made using a 7 nm production process at TSMC. With a die size of 826 mm² and a transistor count of 54,200 million it is a very big chip. GA100 supports DirectX 12 Ultimate (Feature Level 12_2). It features 6912 shading units, 432 texture mapping units and 192 ROPs. Also included are 432 tensor cores which help improve the speed of machine learning applications.
Further reading: Ampere Architecture Whitepaper

Graphics Processor

GPU Name
GA100
Codename
NV170
Architecture
Ampere
Foundry
TSMC
Process Size
7 nm
Transistors
54,200 million
Density
65.6M / mm²
Die Size
826 mm²
Released
Unknown

Graphics Features

DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.2.140
CUDA
8.0
Shader Model
6.5
PureVideo HD
VP11
VDPAU
Feature Set k

Render Config

Shading Units
6912
TMUs
432
ROPs
192
SM Count
108
SFUs
1728
TPCs
54
GPCs
7
Tensor Cores
432
L1 Cache
192 KB per SM
L2 Cache
49152 KB
Max. TDP
400 W

All Ampere GPUs

NVIDIA GPU Architecture History

Graphics cards using the NVIDIA GA100 GPU

Name Chip Memory Shaders TMUs ROPs GPU Clock Memory Clock
NVIDIA Tesla A100 SXM4 40 GB 6912 432 160 1410 MHz 1215 MHz
NVIDIA GRID A100B 48 GB 6912 432 192 900 MHz 1215 MHz
NVIDIA GRID A100A 48 GB 6912 432 192 900 MHz 1215 MHz
NVIDIA GRID A100X 48 GB 6912 432 192 900 MHz 1215 MHz