GeekRadar / The comparison desk

COMPARE.

GPU

NVIDIA

NVIDIA RTX PRO 4000 Blackwell Embedded

Graphics processor

Video memory: 16 GB^*
Memory bus: 256 bit^*
Rated memory bandwidth: 896 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Blackwell 2.0^*
Process node: 5 nm^*
GPU cores: 7,680^*
Graphics interface: PCIe 5.0 x16^*
Reported chip die area: 378 mm²^*
Transistors: 45,600 million^*
GPU L2 cache: 64 MB^*
Graphics processor: GB203^*
Texture mapping units: 240^*
Raster output units: 96^*
Streaming multiprocessors: 60^*
Tensor cores: 240^*
Ray tracing cores: 60^*
Chip foundry: TSMC^*
Card or module format: IGP^*
Declared GPU TDP: 80 W^*
External power connectors: None^*
DirectX / feature level: 12 Ultimate (12_2)^*
OpenGL: 4.6^*
OpenCL: 3.0^*
Vulkan: 1.4^*
CUDA compute capability: 12.0^*
Shader model: 6.9^*

Memory & storage

Memory options: 16^*
Video memory: 16 GB^*
Video memory type: GDDR7^*
Memory bus: 256 bit^*
Rated memory bandwidth: 896 GB/s^*

Power specifications

Thermal design power: 80 W^*

System configuration

Detailed specifications

Specifications 1 · 28 fields

proc transistors: 45,600 million^*
proc die size: 378^* mm²
cores: 7,680^*
tensor cores: 240^*
rt cores: 60^*
sm count: 60^*
base clock: 975^*
boost clock: 1,702^*
memory clock: 1,750^*
ram: 16^*
name: NVIDIA RTX PRO 4000 Blackwell Embedded^*
ram type: GDDR7^*
bus width: 256^*
bus interface: PCIe 5.0 x16^*
ram bandwidth: 896^* GB/s
fp16 performance: 26.14^*
fp32 performance: 26.14^*
fp64 performance: 0.4085^*
tdp: 80^* W
slot width: IGP^*
api cuda: 12^*
is retail board: No^*
gpi value: 15.03^*
vendor: nvd^*
manufacturer: NVIDIA^*
arch name: Blackwell 2.0^*
proc foundry: TSMC^*
proc process size: 5^* nm

Specifications 2 · 39 fields

CUDA compute capability

12.0^*

DirectX / feature level

12 Ultimate (12_2)^*

OpenCL

3.0^*

OpenGL

4.6^*

Shader model

6.9^*

Vulkan

1.4^*

Reported theoretical BF16 peak

52.29^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Graphics interface

PCIe 5.0 x16^*

Memory bus

256^* bit

Reported shader processors

7,680^*

Reported theoretical FP16 peak

26.14^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP32 peak

26.14^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP64 peak

0.4085^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared L1 cache configuration

128 KB (per SM)^*

Declared GPU L2 cache

64 MB^*

Reported display outputs

Portable Device Dependent^*

Theoretical pixel fill rate

163.4 GPixel/s^*

External power connectors

None^*

Reported chip die area

378 mm²^*

Chip foundry

TSMC^*

Process node declaration

5 nm^*

Transistors

45.6 billion^*

Graphics processor

GB203^*

Reported video memory

16^* GB

Rated memory bandwidth

896.0 GB/s^*

Video memory type

GDDR7^*

Raster output units

96^*

Ray tracing cores

60^*

Reported shading units

7,680^*

Card or module format

IGP^*

Streaming multiprocessors

60^*

Declared GPU TDP

80 W^*

Tensor cores

240^*

Theoretical texture fill rate

408.5 GTexel/s^*

Reported theoretical TF32 peak

26.14^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Texture mapping units

240^*

GPU cores

7,680^*

Video memory

16^* GB

Video memory type

GDDR7^*

* Provisional data