Compute Comparison
vs
All comparisons →

Ampere

NVIDIA A10G

wins

3

The NVIDIA A10G is the AWS-specific variant of the A10, available on g5 instances.

AWS inferenceGraphics workloadsMulti-tenant AI

Turing

NVIDIA T4

wins

2

The NVIDIA T4 is the most widely deployed inference GPU in the cloud.

Low-cost inferenceEdge AIMulti-tenant serving

Performance comparison

Visual bars — winner highlighted

NVIDIA A10G

Metric

NVIDIA T4

FP16 TFLOPS

125 T
65 T

Bandwidth

600 GB/s
320 GB/s

VRAM

246 GB
166 GB

TDP (lower=better)

150 W
70 W

TFLOPS/watt

0.83
0.93

GB/s per watt

4.00
4.57

Full specification table

SpecNVIDIA A10GNVIDIA T4
ArchitectureAmpereTuring
VRAM24 GB GDDR616 GB GDDR6
VRAM typeGDDR6GDDR6
Memory bandwidth600 GB/s320 GB/s
FP16 TFLOPS125 TFLOPS65 TFLOPS
TDP150 W70 W
TFLOPS/watt0.83 T/W0.93 T/W
GB/s per watt4.00 GB/s/W4.57 GB/s/W
NVLinkNoNo
Release year20212018

Live cloud pricing

On-demand $/hr across providers — updated in real time

Loading prices…

Power efficiency analysis

TFLOPS/watt and GB/s/watt — critical for data center TCO

NVIDIA A10G

FP16 TFLOPS125 TFLOPS
TDP150 W
TFLOPS/watt0.833 T/W
GB/s per watt4.00 GB/s/W

NVIDIA T4

FP16 TFLOPS65 TFLOPS
TDP70 W
TFLOPS/watt0.929 T/W
GB/s per watt4.57 GB/s/W

TFLOPS/watt measures compute efficiency — how much AI throughput you get per watt of power consumed. For data centers with PUE of 1.2–1.5, a 10% improvement in TFLOPS/watt translates directly to lower electricity costs and cooling requirements. GB/s/watt measures memory bandwidth efficiency, which is the binding constraint for memory-bound LLM inference workloads.

When to choose each GPU

Choose NVIDIA A10G for:

  • AWS inference
  • Graphics workloads
  • Multi-tenant AI

Choose NVIDIA T4 for:

  • Low-cost inference
  • Edge AI
  • Multi-tenant serving