GPU Comparison
GEFORCE
NVIDIA A2
CORE STATE
GA107
VRAM
16 GB
CLOCK SPEED
1770 MHz
TDP
60 W
BUS WIDTH
128 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
2021
VS
GEFORCE
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
—
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
A2
B200
Core Specs
Shading Units
1,280
18,944
+1380.0%
Shaders
1,280
18,944
+1380.0%
TMUs
40
592
+1380.0%
ROPs
32
24
-25.0%
SM Count
10
148
+1380.0%
Clocks
Base Clock
1440 MHz
700 MHz
Boost Clock
1770 MHz
1965 MHz
Memory Clock
1563 MHz
12.5 Gbps effective
2000 MHz
8 Gbps effective
Memory
Memory Size
16 GB
90 GB
VRAM (MB)
16,384
92,160
+462.5%
Memory Type
GDDR6
HBM3e
Memory Bus
128 bit
4096 bit
Bandwidth
200.1 GB/s
4.10 TB/s
Cache
L1 Cache
128 KB (per SM)
256 KB (per SM)
L2 Cache
2 MB
50 MB
Performance
Pixel Rate
56.64 GPixel/s
47.16 GPixel/s
Texture Rate
70.80 GTexel/s
1,163.3 GTexel/s
FP32 (TFLOPS)
4.531 TFLOPS
74.45 TFLOPS
FP64 (TFLOPS)
70.80 GFLOPS (1:64)
37.22 TFLOPS (1:2)
FP16 (TFLOPS)
4.531 TFLOPS (1:1)
1,191.2 TFLOPS (16:1)
AI/RT
RT Cores
10
—
Tensor Cores
40
592
+1380.0%
Power
TDP
60 W
1000 W
TDP (W)
60
1,000
+1566.7%
Suggested PSU
250 W
1400 W
Power Connectors
None
—
Architecture
Architecture
Ampere
Blackwell
GPU Name
GA107
GB100
Generation
Workstation Ampere
(Ax000)
Server Blackwell
(Bxx)
Process Size
8 nm
5 nm
Transistors
8,700 million
104,000 million
Die Size
200 mm²
—
Foundry
Samsung
TSMC
Density
43.5M / mm²
—
API Support
DirectX
12 Ultimate (12_2)
—
OpenGL
4.6
—
Vulkan
1.4
—
OpenCL
3.0
3.0
CUDA
8.6
10.0
Shader Model
6.8
—
Physical
Slot Width
Single-slot
SXM Module
Outputs
No outputs
No outputs
Bus Interface
PCIe 4.0 x8
PCIe 5.0 x16
Other
Production
End-of-life
Active
Predecessor
Quadro Turing
Server Hopper
Successor
Workstation Ada
Server Rubin