NVIDIA GRID RTX T10 4 vs NVIDIA H100 CNX
GPU · Specs Compare
Graphics Card
Release Date
Jan 2020
Mar 2023
Generation
GRID
Tesla Hopper
Type
Professional
AI GPU
Bus Interface
PCIe 3.0 x16
PCIe 5.0 x16
Clock Speeds
Base Clock
1065 MHz
690 MHz
Boost Clock
1395 MHz
1845 MHz
Memory Clock
1750 MHz
1593 MHz
Memory
Memory Size
4GB
80GB
Memory Type
GDDR6
HBM2e
Memory Bus
384bit
5120bit
Bandwidth
672.0GB/s
2039GB/s
Render Config
SM Count
56
114
Shading Units
3584
14592
TMUs
224
456
ROPs
64
24
Tensor Cores
448
456
RT Cores
56
—
L1 Cache
64 KB (per SM)
256 KB (per SM)
L2 Cache
6 MB
50 MB
Theoretical Performance
Pixel Rate
89.28 GPixel/s
44.28 GPixel/s
Texture Rate
312.5 GTexel/s
841.3 GTexel/s
FP16 (half)
20.00 TFLOPS
215.4 TFLOPS
FP32 (float)
9.999 TFLOPS
53.84 TFLOPS
FP64 (double)
312.5 GFLOPS
26.92 TFLOPS
Graphics Processor
GPU Name
TU102
GH100
GPU Variant
TU102-875-A1
—
Architecture
Turing
Hopper
Foundry
TSMC
TSMC
Process Size
12 nm
4 nm
Transistors
18.6 billion
80 billion
Die Size
754 mm²
814 mm²
Board Design
TDP
150W
350W
Suggested PSU
450 W
750 W
Outputs
No outputs
No outputs
Power Connectors
1x 6-pin + 1x 8-pin
8-pin EPS
Graphics Features
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
N/A
OpenCL
3.0
3.0
Vulkan
1.3
N/A
CUDA
7.5
9.0
Shader Model
6.8
N/A
▲ Specs Compare
Data sourced from official specifications and benchmark tests.