GPU architecture	NVIDIA Ampere
CUDA parallelizes the core	6,144
NVIDIA Tensor core	192
NVIDIA RT core	48
Single-precision performance1	19.2 TFLOPS
RT Core performance1	37.4 TFLOPS
Tensor performance1	153.4 TFLOPS2
GPU memory	16 GB GDDR6 with ECC
Memory interface	256-bit
Memory bandwidth	448 GB/s
Maximum power consumption	140 W
Drawing bus	PCI Express 4.0 x 16
Display interfaces	DP 1.4 (4)
Form factor	4.4" H x 9.5" L single slot
Product weight	500 g
Heat dissipation method	Proactive
External power connector	1x 6-pin PCle
Synchronous output	Support for Quadro Sync II