GPU

NVIDIA GeForce RTX 5060

Edit@1 months ago

Intergrated Memory(VRAM)
Capacity

8 GB

(GDDR7 128-bit)

Bandwidth

448 GB/s

64 Token/s

Vector Compute
FP64
299.60 G
FP32
19.18 T
FP16
19.18 T
BF16
19.18 T
INT32
19.18 T
INT8
X

NVIDIA GeForce RTX 5060 General-Purpose Float-Point performance (Vector Performance / Scalar Performance)

FP64: 299.60 GFLOPS

FP32: 19.18 TFLOPS

FP16: 19.18 TFLOPS

BF16: 19.18 TFLOPS

INT32: 19.18 TOPS

Matirx Compute
FP64
X
FP32
X
FP16
38.35 T
76.71 T
FP8
76.71 T
153.42 T
TF32
19.18 T
38.35 T
BF16
38.35 T
76.71 T
INT16
X
INT8
153.42 T
306.83 T
INT4
X
Theoretical peak performance: GeForce-Blackwell

NVIDIA GeForce RTX 5060 AI performance (Tensor Performance / Matrix Performance)

FP16: 38.35 TFLOPS, with sparsity: 76.71 TFLOPS

FP8: 76.71 TFLOPS, with sparsity: 153.42 TFLOPS

TF32: 19.18 TFLOPS, with sparsity: 38.35 TFLOPS

BF16: 38.35 TFLOPS, with sparsity: 76.71 TFLOPS

INT8: 153.42 TOPS, with sparsity: 306.83 TOPS

Hardware Specs
NVIDIA GeForce RTX 5060 is a 4nm chip, has 21900 million transistors, launched by NVIDIA at 2025. It has 8 GB built-in(On-Board/On-Chip) memory with bandwidth up to 448 GB/s. It has 3840 general-purpose ALUs(CUDA cores/Shader cores) and 120 matrix cores(Tensor cores) .
Process Node
4 nm
Launch Year
2025
Vector(CUDA) Cores
3840
Matrix(Tensor) Cores
120
Core Frequency
2280 ~ 2497 MHz
Cache
32MB