NVIDIA RTX 4000 Ada Generation vs NVIDIA B100
We compared two Desktop platform GPUs: 20GB VRAM RTX 4000 Ada Generation and 96GB VRAM B100 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA RTX 4000 Ada Generation 's Advantages
Boost Clock has increased by 18% (2175MHz vs 1837MHz)
Lower TDP (130W vs 1000W)
NVIDIA B100 's Advantages
Released 1 years and 3 months late
More VRAM (96GB vs 20GB)
Larger VRAM bandwidth (4.10TB/s vs 360.0GB/s)
10752 additional rendering cores
Score
Benchmark
FP32 (float)
RTX 4000 Ada Generation
26.73 TFLOPS
B100
+132%
62.08 TFLOPS
Graphics Card
Aug 2023
Release Date
Nov 2024
Workstation Ada
Generation
Server Blackwell
Desktop
Type
Desktop
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
1500 MHz
Base Clock
1665 MHz
2175 MHz
Boost Clock
1837 MHz
2250 MHz
Memory Clock
2000 MHz
Memory
20GB
Memory Size
96GB
GDDR6
Memory Type
HBM3e
160bit
Memory Bus
4096bit
360.0GB/s
Bandwidth
4.10TB/s
Render Config
-
-
-
48
SM Count
132
6144
Shading Units
16896
192
TMUs
528
64
ROPs
24
192
Tensor Cores
528
48
RT Cores
-
128 KB (per SM)
L1 Cache
256 KB (per SM)
48 MB
L2 Cache
50 MB
-
-
-
Theoretical Performance
139.2 GPixel/s
Pixel Rate
44.09 GPixel/s
417.6 GTexel/s
Texture Rate
969.9 GTexel/s
26.73 TFLOPS
FP16 (half)
248.3 TFLOPS
26.73 TFLOPS
FP32 (float)
62.08 TFLOPS
417.6 GFLOPS
FP64 (double)
31.04 TFLOPS
Board Design
130W
TDP
1000W
300 W
Suggested PSU
1400 W
4x DisplayPort 1.4a
Outputs
No outputs
1x 16-pin
Power Connectors
-
Graphics Processor
AD104
GPU Name
GB102
-
-
-
Ada Lovelace
Architecture
Blackwell
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
35.8 billion
Transistors
104 billion
294 mm²
Die Size
Unknown
Graphics Features
12 Ultimate (12_2)
DirectX
N/A
4.6
OpenGL
N/A
3.0
OpenCL
3.0
1.3
Vulkan
N/A
8.9
CUDA
10.1
6.8
Shader Model
N/A
