GPU

NVIDIA GeForce RTX 2050 Laptop

Edit@4 months ago

Intergrated Memory(VRAM)
Capacity

4 GB

(GDDR6 64-bit)

Bandwidth

112 GB/s

Out of Memory Token/s

Vector Compute
FP64
159.40 G
FP32
5.10 T
FP16
5.10 T
BF16
5.10 T
INT32
2.55 T
INT8
X

NVIDIA GeForce RTX 2050 Laptop General-Purpose Float-Point performance (Vector Performance / Scalar Performance)

FP64: 159.40 GFLOPS

FP32: 5.10 TFLOPS

FP16: 5.10 TFLOPS

BF16: 5.10 TFLOPS

INT32: 2.55 TOPS

Matirx Compute
FP64
X
FP32
X
FP16
10.20 T
20.40 T
FP8
X
TF32
5.10 T
10.20 T
BF16
10.20 T
20.40 T
INT16
X
INT8
40.80 T
81.59 T
INT4
81.59 T
163.18 T

NVIDIA GeForce RTX 2050 Laptop AI performance (Tensor Performance / Matrix Performance)

FP16: 10.20 TFLOPS, with sparsity: 20.40 TFLOPS

TF32: 5.10 TFLOPS, with sparsity: 10.20 TFLOPS

BF16: 10.20 TFLOPS, with sparsity: 20.40 TFLOPS

INT8: 40.80 TOPS, with sparsity: 81.59 TOPS

INT4: 81.59 TOPS, with sparsity: 163.18 TOPS

Hardware Specs
NVIDIA GeForce RTX 2050 Laptop is a 8nm chip, has 8700 million transistors, launched by NVIDIA at 2021. It has 4 GB built-in(On-Board/On-Chip) memory with bandwidth up to 112 GB/s. It has 2048 general-purpose ALUs(CUDA cores/Shader cores) and 64 matrix cores(Tensor cores) .
Process Node
8 nm
Launch Year
2021
Vector(CUDA) Cores
2048
Matrix(Tensor) Cores
64
Core Frequency
735 ~ 1245 MHz
Cache
2MB