NVIDIA RTX PRO 4000 Blackwell vs 2x Radeon Pro W6900X
NVIDIA RTX PRO 4000 Blackwell
2x Radeon Pro W6900X
$5,600
|
8960 Shaders
24GB GDDR7
2452MHz
|
2x 5120 Shaders
64GB (2x 32GB) GDDR6
2171MHz
|
|
Peak AI Performance
1.41 PFLOPS
FP4 Tensor Sparse
|
Peak AI Performance
88.92 TFLOPS
FP16
|
|
FP32
43.94 TFLOPS
|
FP32
44.46 TFLOPS
|
|
FP16
43.94 TFLOPS
|
FP16
88.92 TFLOPS
|
|
Form Factor
PCIe Card
1.0-Slots
|
Form Factor
MPX
2x 4.0-Slots
|
|
TDP
140W
|
TDP
600W
|
|
Power Connectors
-
-
-
1x 16-Pin 12VHPWR
|
-
-
-
-
-
|
|
GB6 OpenCL N/A
0%
|
GB6 OpenCL N/A
0%
|
|
GB6 Metal N/A
0%
|
GB6 Metal N/A
0%
|
|
GB6 Vulkan N/A
0%
|
GB6 Vulkan N/A
0%
|
|
GB5 OpenCL N/A
0%
|
GB5 OpenCL 120,660
40%
|
|
GB5 CUDA N/A
0%
|
GB5 CUDA N/A
0%
|
|
GB5 Metal N/A
0%
|
GB5 Metal 154,880
86%
|
|
GB5 Vulkan N/A
0%
|
GB5 Vulkan N/A
0%
|
|
OCT 2020.1 N/A
0%
|
OCT 2020.1 N/A
0%
|
|
OCT Metal N/A
0%
|
OCT Metal 865
147%
|
|
Peak AI
Performance
1.41 PFLOPS
FP4 Tensor Sparse
|
Peak AI
Performance
88.92 TFLOPS
FP16
|
|
FP4
703.04 TFLOPS
Tensor
1.41 PFLOPS
Tensor Sparse
|
-
-
-
|
|
FP8
-
351.52 TFLOPS
Tensor (FP16 Accumulate)
703.04 TFLOPS
Tensor (FP16 Accumulate) Sparse
351.52 TFLOPS
Tensor (FP32 Accumulate)
703.04 TFLOPS
Tensor (FP32 Accumulate) Sparse
|
-
-
-
-
-
-
|
|
FP16
43.94 TFLOPS
175.76 TFLOPS
Tensor (FP16 Accumulate)
351.52 TFLOPS
Tensor (FP16 Accumulate) Sparse
175.76 TFLOPS
Tensor (FP32 Accumulate)
351.52 TFLOPS
Tensor (FP32 Accumulate) Sparse
|
FP16
88.92 TFLOPS
-
-
-
-
|
|
FP32
43.94 TFLOPS
-
-
|
FP32
44.46 TFLOPS
-
-
|
|
FP64
690 GFLOPS
-
|
FP64
2.78 TFLOPS
-
|
|
BF16
43.94 TFLOPS
175.76 TFLOPS
Tensor
351.52 TFLOPS
Tensor Sparse
|
-
-
-
-
|
|
TF32
87.88 TFLOPS
Tensor
175.76 TFLOPS
Tensor Sparse
|
-
-
-
|
|
-
-
-
|
-
-
-
|
|
INT8
-
351.52 TOPS
Tensor
703.04 TOPS
Tensor Sparse
|
-
-
-
-
|
|
INT32
21.97 TOPS
|
-
-
|
|
Ray Tracing
133.3 TOPS
|
-
-
|
|
Pixel
Fillrate
235.392 GPixel/s
|
Pixel
Fillrate
277.888 GPixel/s
|
|
-
-
|
-
-
|
|
Texture
Fillrate
686.56 GTexel/s
|
Texture
Fillrate
694.72 GTexel/s
|
|
Manufacturer
NVIDIA
|
Manufacturer
Apple
|
|
Chip Designer
NVIDIA
|
Chip Designer
AMD
|
|
Architecture
Blackwell
|
Architecture
RDNA 2
|
|
Family
RTX PRO Blackwell
|
Family
Radeon Pro W
|
|
Codename
NV193
GB203
-
-
|
Codename
Sienna Cichlid
Navi 21
Variant
Navi 21 XTA
|
|
Market Segment
Workstation
|
Market Segment
Workstation
|
|
Release Date
3/18/2025
|
Release Date
8/3/2021
|
|
Foundry
TSMC
-
|
Foundry
TSMC
-
|
|
Fabrication Node
4NP
-
|
Fabrication Node
N7
-
|
|
Die Size
378 mm²
-
|
Die Size
2x 520 mm²
-
|
|
Transistor Count
45.6 Billion
-
|
Transistor Count
2x 26.8 Billion
-
|
|
Transistor Density
120.63M/mm²
-
|
Transistor Density
51.56M/mm²
-
|
|
Form
PCIe Card
|
Form
MPX
|
|
Shading Units
8960 Shaders
-
|
Shading Units
2x 5120 Shaders
-
|
|
Texture Mapping Units
280 TMUs
|
Texture Mapping Units
2x 320 TMUs
|
|
Render Output Units
96 ROPs
|
Render Output Units
2x 128 ROPs
|
|
Tensor Cores
280 T-Cores
|
-
-
|
|
Ray-Tracing Cores
70 RT-Cores
|
Ray-Tracing Cores
2x 80 RT-Cores
|
|
Streaming Multiprocessors
70 SMs
|
-
-
|
|
-
-
|
Compute Units
2x 80 CUs
|
|
-
-
|
-
-
|
|
-
-
|
Graphics Processing Clusters
2x 4 GPCs
|
|
-
-
-
2452MHz
|
-
-
1825MHz Base
2171MHz
|
|
-
-
|
L0
32KB/WGP
|
|
L1
64KB/SM Tex
128KB/SM
-
-
|
L1
-
-
-
128KB/Array
|
|
L2
48MB Shared
|
L2
4MB Shared
|
|
-
-
-
|
L3
128MB Shared
1.66TB/s
|
|
24GB
GDDR7
ECC
|
64GB (2x 32GB)
GDDR6
-
|
|
Bus Width
192Bit
|
Bus Width
256Bit
|
|
Clock
1750MHz
Transfer Rate
28GT/s
Bandwidth
672GB/s
|
Clock
2000MHz
Transfer Rate
16GT/s
Bandwidth
512GB/s
|
|
-
-
-
-
-
-
-
-
-
|
-
-
-
-
-
-
-
-
-
|
|
TDP
140W
|
TDP
600W
|
|
-
-
|
Temp
100°C Max
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
4x DisplayPort 2.1
-
-
-
-
-
-
-
-
-
-
-
-
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
2x HDMI 2.0
-
-
8x Thunderbolt 3 (40Gbps)
|
|
Max Resolution
7680x4320
|
Max Resolution
6016x3384
|
|
Max Resolution Refresh Rate
165Hz
|
Max Resolution Refresh Rate
60Hz
|
|
Variable Refresh Rate
G-Sync
FreeSync
-
|
Variable Refresh Rate
-
FreeSync
-
|
|
Display Stream Compression (DSC)
Supported
|
Display Stream Compression (DSC)
Supported
|
|
Multi Monitor Support
4
|
Multi Monitor Support
6
|
|
Content Protection
HDCP 2.3
|
Content Protection
HDCP 2.3
|
|
Model
2x NVENC 9
|
Model
VCN 3.0
|
|
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
|
|
Model
2x NVDEC 6
|
Model
VCN 3.0
|
|
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
|
|
Direct X
12
Direct 3D
12_3
|
Direct X
12
Direct 3D
12_2
|
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
|
OpenGL
4.6
OpenCL
2.1
Vulkan
1.2
|
|
Shader Model
6.8
CUDA
12.8
-
-
PureVideo HD
VP13
VDPAU
Feature Set M
|
Shader Model
6.5
-
-
GFX
10.3
-
-
-
-
|
|
-
-
-
1x Fan
|
-
-
-
-
|
|
Power Connectors
-
-
-
-
-
-
1x 16-Pin 12VHPWR
|
-
-
-
-
-
-
-
-
|
|
Slots Required
1.0
PCIe Version
5.0
PCIe Lanes
16
|
Slots Required
4.0
PCIe Version
4.0
PCIe Lanes
16
|
|
Multi GPU Support
Supported
Type
NVLink
|
Multi GPU Support
Supported
Type
Infinity Fabric
|
|
Height
111.8 mm (4.4 in)
Width
241.3 mm (9.5 in)
Depth
20 mm (0.79 in)
|
-
-
-
-
-
-
|
Copy Link