20 GB

GDDR6

1635 MHz

Boost Clock

150 W

TDP

8 nm

Process
Share on Social Media
Share A10M specifications with others

Technical Specifications

Core Specifications
Architecture: Ampere
Process Size: 8 nm
Base Clock: 975 MHz
Boost Clock: 1635 MHz
Shader Units: 7168
SMs: 56
Tensor Cores: 224
RT Cores: 56
TMUs: 224
ROPs: 80
Memory & Performance
Memory Size: 20 GB
Memory Type: GDDR6
Memory Bandwidth: 500.2 GB/s
Bus Interface: PCIe 4.0 x16
TDP: 150 W
Suggested PSU: 450 W
Cache & Die Information
L1 Cache: 128 KB
L2 Cache: 6.0 MB
Die Size: 628 mm²
Transistors: 28.3 billion
Transistor Density: 45.1 million/mm²

API Support

DirectX
12.2
OpenGL
4.6
Vulkan
1.3
Cuda Compute
8.6 sm_86

Physical Specifications

Board Length: 267 mm
Board Width: 112 mm
Power Connectors: 8-pin EPS

Quick Facts

Release Date: 2022-02-01
Manufacturer: NVIDIA
Generation: Server Ampere(Axx)

Performance

Pixel Fill Rate 130.8 GPixel/s
Texture Fill Rate 366.2 GTexel/s
FP32 Performance 23.4 TFLOPS

CUDA Toolkit Compatibility

Cuda Compute
8.6 sm_86
Architecture

Ampere

Supported CUDA Toolkits
11.1 11.2 11.3 11.4 11.5 11.6 11.7 11.8 12.0 12.1 12.2 12.3 12.4 12.5 12.6 12.7 12.8

Ampere architecture with 2nd gen RT cores and hardware AV1 decode