GPU Comparison
GEFORCE
NVIDIA GB10
CORE STATE
GB20B
VRAM
128 GB
CLOCK SPEED
2418 MHz
TDP
140 W
BUS WIDTH
256 bit
ARCHITECTURE
Blackwell 2.0
PROCESS
5 nm
LAUNCH DATE
2025
VS
GEFORCE
Quadro K620M
CORE STATE
GM108S
VRAM
2 GB
CLOCK SPEED
1124 MHz
TDP
30 W
BUS WIDTH
64 bit
ARCHITECTURE
Maxwell
PROCESS
28 nm
LAUNCH DATE
2015
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
GB10
Quadro K620M
Core Specs
Shading Units
6,144
384
-93.8%
Shaders
6,144
384
-93.8%
TMUs
384
16
-95.8%
ROPs
48
8
-83.3%
SM Count
48
—
Clocks
Base Clock
1665 MHz
1029 MHz
Boost Clock
2418 MHz
1124 MHz
Memory Clock
1067 MHz
8.5 Gbps effective
1001 MHz
2 Gbps effective
Memory
Memory Size
128 GB
2 GB
VRAM (MB)
131,072
2,048
-98.4%
Memory Type
LPDDR5X
DDR3
Memory Bus
256 bit
64 bit
Bandwidth
273.2 GB/s
16.02 GB/s
Cache
L1 Cache
256 KB (per SM)
64 KB (per SMM)
L2 Cache
50 MB
1024 KB
Performance
Pixel Rate
116.1 GPixel/s
8.992 GPixel/s
Texture Rate
928.5 GTexel/s
17.98 GTexel/s
FP32 (TFLOPS)
29.71 TFLOPS
863.2 GFLOPS
FP64 (TFLOPS)
464.3 GFLOPS (1:64)
26.98 GFLOPS (1:32)
FP16 (TFLOPS)
29.71 TFLOPS (1:1)
—
AI/RT
RT Cores
48
—
Tensor Cores
384
—
Power
TDP
140 W
30 W
TDP (W)
140
30
-78.6%
Suggested PSU
300 W
—
Power Connectors
None
None
Architecture
Architecture
Blackwell 2.0
Maxwell
GPU Name
GB20B
GM108S
Generation
Server Blackwell
(Bxx)
Quadro Kepler-M
(Kx200M)
Process Size
5 nm
28 nm
Transistors
—
1,020 million
Die Size
—
77 mm²
Foundry
TSMC
TSMC
Density
—
13.2M / mm²
API Support
DirectX
—
12 (11_0)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
12.1
5.0
Shader Model
—
6.7 (5.1)
Physical
Slot Width
IGP
MXM Module
Length
150 mm
5.9 inches
—
Height
51 mm
2 inches
—
Outputs
1x HDMI
Portable Device Dependent
Bus Interface
PCIe 5.0 x16
MXM-A (3.0)
Other
Launch Price
3,999 USD
—
Production
Active
End-of-life
Predecessor
Server Hopper
Quadro Fermi-M
Successor
Server Rubin
Quadro Maxwell-M