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
P106-100
CORE STATE
GP106
VRAM
6 GB
CLOCK SPEED
1709 MHz
TDP
120 W
BUS WIDTH
192 bit
ARCHITECTURE
Pascal
PROCESS
16 nm
LAUNCH DATE
2017
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
3dmark_3dmark_steel_nomad_dx12
DETAILED SPECIFICATIONS
SPECIFICATION
GB10
P106-100
Core Specs
Shading Units
6,144
1,280
-79.2%
Shaders
6,144
1,280
-79.2%
TMUs
384
80
-79.2%
ROPs
48
48
0.0%
SM Count
48
10
-79.2%
Clocks
Base Clock
1665 MHz
1506 MHz
Boost Clock
2418 MHz
1709 MHz
Memory Clock
1067 MHz
8.5 Gbps effective
2002 MHz
8 Gbps effective
Memory
Memory Size
128 GB
6 GB
VRAM (MB)
131,072
6,144
-95.3%
Memory Type
LPDDR5X
GDDR5
Memory Bus
256 bit
192 bit
Bandwidth
273.2 GB/s
192.2 GB/s
Cache
L1 Cache
256 KB (per SM)
48 KB (per SM)
L2 Cache
50 MB
1536 KB
Performance
Pixel Rate
116.1 GPixel/s
82.03 GPixel/s
Texture Rate
928.5 GTexel/s
136.7 GTexel/s
FP32 (TFLOPS)
29.71 TFLOPS
4.375 TFLOPS
FP64 (TFLOPS)
464.3 GFLOPS (1:64)
136.7 GFLOPS (1:32)
FP16 (TFLOPS)
29.71 TFLOPS (1:1)
68.36 GFLOPS (1:64)
AI/RT
RT Cores
48
—
Tensor Cores
384
—
Power
TDP
140 W
120 W
TDP (W)
140
120
-14.3%
Suggested PSU
300 W
300 W
Power Connectors
None
1x 6-pin
Architecture
Architecture
Blackwell 2.0
Pascal
GPU Name
GB20B
GP106
Generation
Server Blackwell
(Bxx)
Mining GPUs
Process Size
5 nm
16 nm
Transistors
—
4,400 million
Die Size
—
200 mm²
Foundry
TSMC
TSMC
Density
—
22.0M / mm²
API Support
DirectX
—
12 (12_1)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
12.1
6.1
Shader Model
—
6.8
Physical
Slot Width
IGP
Dual-slot
Length
150 mm
5.9 inches
250 mm
9.8 inches
Height
51 mm
2 inches
—
Outputs
1x HDMI
No outputs
Bus Interface
PCIe 5.0 x16
PCIe 1.0 x16
Other
Launch Price
3,999 USD
—
Production
Active
End-of-life
Predecessor
Server Hopper
—
Successor
Server Rubin
—