NVIDIA GRID A100A vs NVIDIA GRID RTX T10-8 Full Specs

Summary

6,912 Shaders 900MHz	4,608 Shaders 1.06GHz
32GB HBM2e1.87TB/s	8GB GDDR6672GB/s
Peak·FP32·FP16 12.44 TFLOPS	Peak·FP32·FP16 9.81 TFLOPS
Form Factor vGPU	Form Factor PCIe Card
TDP 400W	TDP 260W
Power Connectors -	Power Connectors 1x 8-Pin 1x 6-Pin

Highlights

Peak OPS

Peak OPS FP32 FP16 Tensor FP16-16 FP16-16 Tensor Sparse Tensor FP16-32 FP16-32 Tensor Sparse FP64 Tensor FP64 BF16 Tensor BF16 BF16 Tensor Sparse Tensor TF32 Tensor INT8 Tensor INT4 Memory Bandwidth TDP

GRID A100A1.59 POPSINT4 Tensor Sparse

x5.07

GRID RTX T10-8314.1 TOPSINT4 Tensor

x1

Show Top Score

Theoretical Performance

Clock Speed900MHz UC·Boost·OC	Clock Speed1065MHz UC·Boost·OC
Peak OPS 1.59 POPSINT4 Tensor Sparse	Peak OPS 314.1 TOPSINT4 Tensor
Tensor FP16-16 199.1 TFLOPS FP16-16 Tensor Sparse 398.1 TFLOPS Tensor FP16-32 199.1 TFLOPS FP16-32 Tensor Sparse 398.1 TFLOPS	Tensor FP16-16 78.52 TFLOPS FP16-16 Tensor Sparse - Tensor FP16-32 39.26 TFLOPS FP16-32 Tensor Sparse -
BF16 24.88 TFLOPS Tensor BF16 199.1 TFLOPS BF16 Tensor Sparse 398.1 TFLOPS	BF16 - Tensor BF16 39.26 TFLOPS BF16 Tensor Sparse -
Tensor TF32 99.53 TFLOPS	Tensor TF32 39.26 TFLOPS
FP32 12.44 TFLOPS	FP32 9.81 TFLOPS
FP64 6.22 TFLOPS Tensor FP64 12.44 TFLOPS	FP64 306.7 GFLOPS Tensor FP64 -
Tensor INT4 796.3 TOPS	Tensor INT4 314.1 TOPS
Tensor INT8 398.1 TOPS	Tensor INT8 157 TOPS
Ray -	Ray 29.61 TOPS
Pixel Rate 172.8 GPixel/s	Pixel Rate 102.2 GPixel/s
Texture Rate 388.8 GTexel/s	Texture Rate 306.7 GTexel/s

Core Config

Shaders 6,912 Shaders	Shaders 4,608 Shaders
TMUs 432 TMUs	TMUs 288 TMUs
ROPs 192 ROPs	ROPs 96 ROPs
Tensor Cores 432 T-Cores	Tensor Cores 576 T-Cores
RT Cores -	RT Cores 72 RT-Cores
SMs 108 SMs	SMs 72 SMs

Clocks

Boost Clock 900MHz

Boost Clock 1.06GHz

Cache

L2 Cache 41MB shared

L2 Cache 6.1MB shared

Memory

32GB HBM2e	8GB GDDR6
Memory Bus 6144-bit	Memory Bus 384-bit
Memory Speed 2.4GT/s	Memory Speed 14GT/s
Memory Bandwidth 1.87TB/s	Memory Bandwidth 672GB/s

Power & Thermals

TDP 400W

TDP 260W

Encoder

-	Encoder Model NVENC 7

Decoder

Decoder Model 5x NVDEC 4

Decoder Model NVDEC 4

Card

Form Factor vGPU	Form Factor PCIe Card
PCIe -	PCIe 2-Slots
-	Height 111 mm (4.37") Width 267 mm (10.51") Depth 40 mm (1.57")
Cooling Passive	Cooling Passive
Power Connectors -	Power Connectors 1x 8-Pin, 1x 6-Pin

Chip

Manufacturer NVIDIA	Manufacturer NVIDIA
Chip Designer NVIDIA	Chip Designer NVIDIA
Architecture Ampere	Architecture Turing
Family GRID	Family GRID
Branding	Branding
Codename NV170	Codename NV162
Chip Variant -	Chip Variant TU102-300A-K1-A1
Market Segment Server	Market Segment Server
Release Date May 14, 2020	Release Date Jan 1, 2020

Fabrication

Foundry TSMC	Foundry TSMC
Fabrication Node 7N	Fabrication Node 12FFN
Die Size 826mm²	Die Size 754mm²
Transistor Count 54.2 Billion	Transistor Count 18.6 Billion
Transistor Density 65.62 MTr/mm²	Transistor Density 24.67 MTr/mm²

Images

No images available

No images available

Competitors

NVIDIAGRID A100A vs GPU AMDInstinct MI300P NVIDIAGRID A100A vs GPU NVIDIAGeForce RTX 5080

NVIDIAGRID RTX T10-8 vs GPU AMDRadeon RX 9070 XT NVIDIAGRID RTX T10-8 vs GPU AMDRadeon RX 7800 XT NVIDIAGRID RTX T10-8 vs GPU AMDRadeon AI PRO R9700 Show 12 more