GPU

NVIDIA GeForce RTX 5080

Edit@1 months ago

Intergrated Memory(VRAM)
Capacity

16 GB

(GDDR7 256-bit)

Bandwidth

960 GB/s

137 Token/s

Vector Compute
FP64
879.30 G
FP32
56.28 T
FP16
56.28 T
BF16
56.28 T
INT32
56.28 T
INT8
X

NVIDIA GeForce RTX 5080 General-Purpose Float-Point performance (Vector Performance / Scalar Performance)

FP64: 879.30 GFLOPS

FP32: 56.28 TFLOPS

FP16: 56.28 TFLOPS

BF16: 56.28 TFLOPS

INT32: 56.28 TOPS

Matirx Compute
FP64
X
FP32
X
FP16
112.55 T
225.10 T
FP8
225.10 T
450.21 T
TF32
56.28 T
112.55 T
BF16
112.55 T
225.10 T
INT16
X
INT8
450.21 T
900.42 T
INT4
X

NVIDIA GeForce RTX 5080 AI performance (Tensor Performance / Matrix Performance)

FP16: 112.55 TFLOPS, with sparsity: 225.10 TFLOPS

FP8: 225.10 TFLOPS, with sparsity: 450.21 TFLOPS

TF32: 56.28 TFLOPS, with sparsity: 112.55 TFLOPS

BF16: 112.55 TFLOPS, with sparsity: 225.10 TFLOPS

INT8: 450.21 TOPS, with sparsity: 900.42 TOPS

Hardware Specs
NVIDIA GeForce RTX 5080 is a 5nm chip, has 45600 million transistors, launched by NVIDIA at 2025. It has 16 GB built-in(On-Board/On-Chip) memory with bandwidth up to 960 GB/s. It has 10752 general-purpose ALUs(CUDA cores/Shader cores) and 336 matrix cores(Tensor cores) .
Process Node
5 nm
Launch Year
2025
Vector(CUDA) Cores
10752
Matrix(Tensor) Cores
336
Core Frequency
2295 ~ 2617 MHz
Cache
64MB