GPU

NVIDIA RTX A2000

Edit@2 months ago

Intergrated Memory(VRAM)
Capacity

6 GB

(GDDR6 192-bit)

Bandwidth

288 GB/s

Out of Memory Token/s

Vector Compute
FP64
124.80 G
FP32
7.99 T
FP16
7.99 T
BF16
NI
INT32
NI
INT8
X

NVIDIA RTX A2000 General-Purpose Float-Point performance (Vector Performance / Scalar Performance)

FP64: 124.80 GFLOPS

FP32: 7.99 TFLOPS

FP16: 7.99 TFLOPS

Matirx Compute
FP64
X
FP32
X
FP16
31.95 T
63.90 T
FP8
X
TF32
15.97 T
31.95 T
BF16
31.95 T
63.90 T
INT16
X
INT8
63.90 T
127.80 T
INT4
127.80 T
255.59 T

NVIDIA RTX A2000 AI performance (Tensor Performance / Matrix Performance)

FP16: 31.95 TFLOPS, with sparsity: 63.90 TFLOPS

TF32: 15.97 TFLOPS, with sparsity: 31.95 TFLOPS

BF16: 31.95 TFLOPS, with sparsity: 63.90 TFLOPS

INT8: 63.90 TOPS, with sparsity: 127.80 TOPS

INT4: 127.80 TOPS, with sparsity: 255.59 TOPS

Hardware Specs
NVIDIA RTX A2000 is a 8nm chip, has 12000 million transistors, launched by NVIDIA at 2021. It has 6 GB built-in(On-Board/On-Chip) memory with bandwidth up to 288 GB/s. It has 3328 general-purpose ALUs(CUDA cores/Shader cores) and 104 matrix cores(Tensor cores) .
Process Node
8 nm
Launch Year
2021
Vector(CUDA) Cores
3328
Matrix(Tensor) Cores
104
Core Frequency
562 ~ 1200 MHz
Cache
3MB

Comment without registration

Share your experience with NVIDIA RTX A2000 / Found an Error? Help Us Improve!