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 RTX 5070
CORE STATE
GB205
VRAM
12 GB
CLOCK SPEED
2512 MHz
TDP
250 W
BUS WIDTH
192 bit
ARCHITECTURE
Blackwell 2.0
PROCESS
5 nm
LAUNCH DATE
2025
PERFORMANCE BENCHMARKS
geekbench_opencl
3dmark_3dmark_steel_nomad_dx12
geekbench_vulkan
passmark_directx_10
passmark_directx_11
passmark_directx_12
passmark_directx_9
passmark_g2d
passmark_g3d
passmark_gpu_compute
DETAILED SPECIFICATIONS
SPECIFICATION
B200
RTX 5070
Core Specs
Shading Units
18,944
6,144
-67.6%
Shaders
18,944
6,144
-67.6%
TMUs
592
192
-67.6%
ROPs
24
80
+233.3%
SM Count
148
48
-67.6%
Clocks
Base Clock
700 MHz
2325 MHz
Boost Clock
1965 MHz
2512 MHz
Memory Clock
2000 MHz
8 Gbps effective
1750 MHz
28 Gbps effective
Memory
Memory Size
90 GB
12 GB
VRAM (MB)
92,160
12,288
-86.7%
Memory Type
HBM3e
GDDR7
Memory Bus
4096 bit
192 bit
Bandwidth
4.10 TB/s
672.0 GB/s
Cache
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
50 MB
48 MB
Performance
Pixel Rate
47.16 GPixel/s
201.0 GPixel/s
Texture Rate
1,163.3 GTexel/s
482.3 GTexel/s
FP32 (TFLOPS)
74.45 TFLOPS
30.87 TFLOPS
FP64 (TFLOPS)
37.22 TFLOPS (1:2)
482.3 GFLOPS (1:64)
FP16 (TFLOPS)
1,191.2 TFLOPS (16:1)
30.87 TFLOPS (1:1)
AI/RT
RT Cores
—
48
Tensor Cores
592
192
-67.6%
Power
TDP
1000 W
250 W
TDP (W)
1,000
250
-75.0%
Suggested PSU
1400 W
600 W
Power Connectors
—
1x 16-pin
Architecture
Architecture
Blackwell
Blackwell 2.0
GPU Name
GB100
GB205
Generation
Server Blackwell
(Bxx)
GeForce 50
Process Size
5 nm
5 nm
Transistors
104,000 million
31,100 million
Die Size
—
263 mm²
Foundry
TSMC
TSMC
Density
—
118.3M / mm²
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
10.0
12.0
Shader Model
—
6.8
Physical
Slot Width
SXM Module
Dual-slot
Length
—
245 mm
9.6 inches
Height
—
115 mm
4.5 inches
Outputs
No outputs
1x HDMI 2.1b3x DisplayPort 2.1b
Bus Interface
PCIe 5.0 x16
PCIe 5.0 x16
Other
Launch Price
—
549 USD
Production
Active
Active
Predecessor
Server Hopper
GeForce 40
Successor
Server Rubin
—