GPU Comparison
GEFORCE
NVIDIA B300 SXM6 AC
CORE STATE
GB110
VRAM
288 GB
CLOCK SPEED
2032 MHz
TDP
1100 W
BUS WIDTH
8192 bit
ARCHITECTURE
Blackwell Ultra
PROCESS
5 nm
LAUNCH DATE
2025
VS
GEFORCE
RTX A2000
CORE STATE
GA106
VRAM
6 GB
CLOCK SPEED
1200 MHz
TDP
70 W
BUS WIDTH
192 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
2021
PERFORMANCE BENCHMARKS
geekbench_opencl
3dmark_3dmark_steel_nomad_dx12
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
B300 SXM6 AC
RTX A2000
Core Specs
Shading Units
18,944
3,328
-82.4%
Shaders
18,944
3,328
-82.4%
TMUs
592
104
-82.4%
ROPs
24
48
+100.0%
SM Count
148
26
-82.4%
Clocks
Base Clock
1665 MHz
562 MHz
Boost Clock
2032 MHz
1200 MHz
Memory Clock
2000 MHz 8 Gbps effective
1500 MHz
12 Gbps effective
Memory
Memory Size
288 GB
6 GB
VRAM (MB)
294,912
6,144
-97.9%
Memory Type
HBM3e
GDDR6
Memory Bus
8192 bit
192 bit
Bandwidth
8.19 TB/s
288.0 GB/s
Cache
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
126 MB
3 MB
Performance
Pixel Rate
48.77 GPixel/s
57.60 GPixel/s
Texture Rate
1,202.9 GTexel/s
124.8 GTexel/s
FP32 (TFLOPS)
76.99 TFLOPS
7.987 TFLOPS
FP64 (TFLOPS)
1,202.9 GFLOPS (1:64)
124.8 GFLOPS (1:64)
FP16 (TFLOPS)
76.99 TFLOPS (1:1)
7.987 TFLOPS (1:1)
AI/RT
RT Cores
—
26
Tensor Cores
592
104
-82.4%
Power
TDP
1100 W
70 W
TDP (W)
1,100
70
-93.6%
Suggested PSU
1500 W
250 W
Power Connectors
—
None
Architecture
Architecture
Blackwell Ultra
Ampere
GPU Name
GB110
GA106
Generation
Server Blackwell (Bxx)
Workstation Ampere
(Ax000)
Process Size
5 nm
8 nm
Transistors
208,000 million
12,000 million
Die Size
1628 mm²
276 mm²
Foundry
TSMC
Samsung
Density
127.8M / mm²
43.5M / mm²
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
10.3
8.6
Shader Model
—
6.8
Physical
Slot Width
SXM Module
Dual-slot
Length
—
167 mm
6.6 inches
Height
—
69 mm
2.7 inches
Outputs
No outputs
4x mini-DisplayPort 1.4a
Bus Interface
PCIe 6.0 x16
PCIe 4.0 x16
Other
Launch Price
—
449 USD
Production
Active
End-of-life
Predecessor
Server Hopper
Quadro Turing
Successor
Server Rubin
Workstation Ada