NVIDIA AD103 Graphics Processing Unit (GPU)

Specs based on NVIDIA GeForce RTX 4080 SUPER

Most Powerful Variant

Cores : TMUs : ROPs

10240 : 304 : 112

Typical Memory Size

Typical Memory Bus

AD103 GPU

Manufacturer

NVIDIA

Release Date

January 31st, 2024

Architecture

Ada

Fabrication Process

4 nm

Die Size

379 mm²

Transistors Count

45.9B

Transistors Density

121.1M ^TRAN/_mm²

Tensor Cores

320

RT Cores

80

CUDAs

10240

SM

80

TMUs

304

ROPs

112

Base Clock

2295 MHz

Memory Size

16384 MB

Memory Type

GDDR6X

Memory Bus Width

256-bit

Memory Bandwidth

736.0 GB/s

Pixel Fillrate

285.6 GPixels/s

Texture Fillrate

775.2 GTexel/s

SPFP Performance

52.2 TFLOPS

Performance per W

163.2 GFLOPS/W

Performance per mm²

137.8 GFLOPS/mm²

Interface

PCI-Express 4.0 x16

Typical Thermal Design Power

320 W

SLI Support

Unsupported

DirectX

12.2

Vulkan

1.2

OpenGL

4.6

OpenCL

3.0

Pictures

Front

AD103 GPU Cards5

	Model	GPU Model	Unified Cores	Boost Clock	Memory Clock	Memory Config.
	NVIDIA GeForce RTX 4080 SUPER	AD103-400	10240	2550 MHz	23 Gbps	16 GB G6X 256b
	NVIDIA GeForce RTX 4080	AD103-300	9728	2505 MHz	22 Gbps	16 GB G6X 256b
	NVIDIA GeForce RTX 4090 Laptop GPU	X21-X11	9728	2040 MHz	18 Gbps	16 GB G6 256b
	NVIDIA RTX A5000 ADA Laptop GPU	AD103	9728	2190 MHz	18 Gbps	16 GB G6 256b
	NVIDIA GeForce RTX 4070 Ti SUPER	AD103-275	8448	2610 MHz	21 Gbps	16 GB G6X 256b