GPU Comparison
GEFORCE
NVIDIA B200
CORE STATE
GB100
VRAM
90 GB
CLOCK SPEED
1965 MHz
TDP
1000 W
BUS WIDTH
4096 bit
ARCHITECTURE
Blackwell
PROCESS
5 nm
LAUNCH DATE
—
VS
GEFORCE
Quadro K4000M
CORE STATE
GK104
VRAM
4 GB
CLOCK SPEED
601 MHz
TDP
100 W
BUS WIDTH
256 bit
ARCHITECTURE
Kepler
PROCESS
28 nm
LAUNCH DATE
2012
PERFORMANCE BENCHMARKS
geekbench_opencl
DETAILED SPECIFICATIONS
SPECIFICATION
B200
Quadro K4000M
Core Specs
Shading Units
18,944
960
-94.9%
Shaders
18,944
960
-94.9%
TMUs
592
80
-86.5%
ROPs
24
32
+33.3%
SM Count
148
—
Clocks
Base Clock
700 MHz
601 MHz
Boost Clock
1965 MHz
601 MHz
Memory Clock
2000 MHz
8 Gbps effective
700 MHz
2.8 Gbps effective
Memory
Memory Size
90 GB
4 GB
VRAM (MB)
92,160
4,096
-95.6%
Memory Type
HBM3e
GDDR5
Memory Bus
4096 bit
256 bit
Bandwidth
4.10 TB/s
89.60 GB/s
Cache
L1 Cache
256 KB (per SM)
16 KB (per SMX)
L2 Cache
50 MB
512 KB
Performance
Pixel Rate
47.16 GPixel/s
12.02 GPixel/s
Texture Rate
1,163.3 GTexel/s
48.08 GTexel/s
FP32 (TFLOPS)
74.45 TFLOPS
1,153.9 GFLOPS
FP64 (TFLOPS)
37.22 TFLOPS (1:2)
48.08 GFLOPS (1:24)
FP16 (TFLOPS)
1,191.2 TFLOPS (16:1)
—
AI/RT
Tensor Cores
592
—
Power
TDP
1000 W
100 W
TDP (W)
1,000
100
-90.0%
Suggested PSU
1400 W
—
Power Connectors
—
None
Architecture
Architecture
Blackwell
Kepler
GPU Name
GB100
GK104
Generation
Server Blackwell
(Bxx)
Quadro Kepler-M
(Kx000M)
Process Size
5 nm
28 nm
Transistors
104,000 million
3,540 million
Die Size
—
294 mm²
Foundry
TSMC
TSMC
Density
—
12.0M / mm²
API Support
DirectX
—
12 (11_0)
OpenGL
—
4.6
Vulkan
—
1.2.175
OpenCL
3.0
3.0
CUDA
10.0
3.0
Shader Model
—
6.5 (5.1)
Physical
Slot Width
SXM Module
MXM Module
Outputs
No outputs
Portable Device Dependent
Bus Interface
PCIe 5.0 x16
MXM-B (3.0)
Other
Production
Active
End-of-life
Predecessor
Server Hopper
Quadro Fermi-M
Successor
Server Rubin
Quadro Maxwell-M