AMD Radeon PRO W7900 vs NVIDIA B300

GPU · Specs Compare

Products
GPU AMD Radeon PRO W7900 AMD · 2023-04
GPU NVIDIA B300 NVIDIA · 2025-09
Graphics Card
Release Date
May 2023
Sep 2025
Generation
Radeon Pro Navi
Server Blackwell
Type
Professional
Desktop
Bus Interface
PCIe 4.0 x16
PCIe 5.0 x16
Clock Speeds
Base Clock
1760 MHz
1665 MHz
Boost Clock
2495 MHz
2032 MHz
Memory Clock
2250 MHz
2000 MHz
Memory
Memory Size
48GB
144GB
Memory Type
GDDR6
HBM3e
Memory Bus
384bit
4096bit
Bandwidth
864.0GB/s
4.10TB/s
Render Config
Compute Units
96
—
Shading Units
6144
18944
TMUs
384
592
ROPs
192
24
RT Cores
96
—
L1 Cache
256 KB per Array
256 KB (per SM)
L2 Cache
6 MB
50 MB
L3 Cache
96 MB
—
SM Count
—
148
Tensor Cores
—
592
Theoretical Performance
Pixel Rate
479.0 GPixel/s
48.77 GPixel/s
Texture Rate
958.1 GTexel/s
1202.9 GTexel/s
FP16 (half)
122.6 TFLOPS
1231.8 TFLOPS
FP32 (float)
61.32 TFLOPS
76.99 TFLOPS
FP64 (double)
1.916 TFLOPS
1202.9 GFLOPS
Graphics Processor
GPU Name
Navi 31
GB110
GPU Variant
Navi 31 XTW
—
Architecture
RDNA 3.0
Blackwell Ultra
Foundry
TSMC
TSMC
Process Size
5 nm
5 nm
Transistors
57.7 billion
104 billion
Die Size
529 mm²
Unknown
Board Design
TDP
295W
1400W
Suggested PSU
600 W
1800 W
Outputs
3x DisplayPort 2.11x mini-DisplayPort 2.1
No outputs
Power Connectors
2x 8-pin
—
Graphics Features
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
N/A
OpenCL
2.0
3.0
Vulkan
1.4
N/A
Shader Model
6.9
N/A
CUDA
—
10.3
▲ Specs Compare Data sourced from official specifications and benchmark tests.