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
GeForce MX350
CORE STATE
GP107S
VRAM
2 GB
CLOCK SPEED
1468 MHz
TDP
20 W
BUS WIDTH
64 bit
ARCHITECTURE
Pascal
PROCESS
14 nm
LAUNCH DATE
2020
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
B200
MX350
Core Specs
Shading Units
18,944
640
-96.6%
Shaders
18,944
640
-96.6%
TMUs
592
32
-94.6%
ROPs
24
16
-33.3%
SM Count
148
5
-96.6%
Clocks
Base Clock
700 MHz
1354 MHz
Boost Clock
1965 MHz
1468 MHz
Memory Clock
2000 MHz
8 Gbps effective
1752 MHz
7 Gbps effective
Memory
Memory Size
90 GB
2 GB
VRAM (MB)
92,160
2,048
-97.8%
Memory Type
HBM3e
GDDR5
Memory Bus
4096 bit
64 bit
Bandwidth
4.10 TB/s
56.06 GB/s
Cache
L1 Cache
256 KB (per SM)
48 KB (per SM)
L2 Cache
50 MB
512 KB
Performance
Pixel Rate
47.16 GPixel/s
23.49 GPixel/s
Texture Rate
1,163.3 GTexel/s
46.98 GTexel/s
FP32 (TFLOPS)
74.45 TFLOPS
1.879 TFLOPS
FP64 (TFLOPS)
37.22 TFLOPS (1:2)
58.72 GFLOPS (1:32)
FP16 (TFLOPS)
1,191.2 TFLOPS (16:1)
29.36 GFLOPS (1:64)
AI/RT
Tensor Cores
592
—
Power
TDP
1000 W
20 W
TDP (W)
1,000
20
-98.0%
Suggested PSU
1400 W
—
Power Connectors
—
None
Architecture
Architecture
Blackwell
Pascal
GPU Name
GB100
GP107S
Generation
Server Blackwell
(Bxx)
GeForce MX
(3xx)
Process Size
5 nm
14 nm
Transistors
104,000 million
3,300 million
Die Size
—
132 mm²
Foundry
TSMC
Samsung
Density
—
25.0M / mm²
API Support
DirectX
—
12 (12_1)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
10.0
6.1
Shader Model
—
6.8
Physical
Slot Width
SXM Module
—
Outputs
No outputs
Portable Device Dependent
Bus Interface
PCIe 5.0 x16
PCIe 3.0 x4
Other
Production
Active
End-of-life
Predecessor
Server Hopper
—
Successor
Server Rubin
—