NVIDIA L40 CNX vs NVIDIA B200 SXM 192 GB
GPU · Specs Compare
Graphics Card
Release Date
Oct 2022
Jan 2024
Generation
Tesla Ada
Server Blackwell
Type
Professional
Desktop
Bus Interface
PCIe 4.0 x16
PCIe 5.0 x16
Clock Speeds
Base Clock
1005 MHz
1665 MHz
Boost Clock
2475 MHz
1837 MHz
Memory Clock
2250 MHz
2000 MHz
Memory
Memory Size
24GB
96GB
Memory Type
GDDR6
HBM3e
Memory Bus
384bit
4096bit
Bandwidth
864.0GB/s
4.10TB/s
Render Config
SM Count
142
132
Shading Units
18176
16896
TMUs
568
528
ROPs
192
24
Tensor Cores
568
528
RT Cores
142
—
L1 Cache
128 KB (per SM)
256 KB (per SM)
L2 Cache
48 MB
50 MB
Theoretical Performance
Pixel Rate
475.2 GPixel/s
44.09 GPixel/s
Texture Rate
1406 GTexel/s
969.9 GTexel/s
FP16 (half)
89.97 TFLOPS
248.3 TFLOPS
FP32 (float)
89.97 TFLOPS
62.08 TFLOPS
FP64 (double)
1406 GFLOPS
31.04 TFLOPS
Graphics Processor
GPU Name
AD102
GB100
GPU Variant
AD102-???-A1
—
Architecture
Ada Lovelace
Blackwell
Foundry
TSMC
TSMC
Process Size
4 nm
5 nm
Transistors
76.3 billion
104 billion
Die Size
608 mm²
Unknown
Board Design
TDP
300W
1000W
Suggested PSU
700 W
1400 W
Outputs
1x HDMI 2.13x DisplayPort 1.4a
No outputs
Power Connectors
1x 16-pin
—
Graphics Features
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
N/A
OpenCL
3.0
3.0
Vulkan
1.3
N/A
CUDA
8.9
10.0
Shader Model
6.6
N/A
▲ Specs Compare
Data sourced from official specifications and benchmark tests.