Compute Comparison
vs
All comparisons →

Ada Lovelace

NVIDIA L4

wins

3

The NVIDIA L4 is the most power-efficient Ada Lovelace GPU at just 72W TDP.

Efficient inferenceVideo transcodingEdge AI

Ampere

NVIDIA A10

wins

1

The NVIDIA A10 is a versatile Ampere GPU with 24GB GDDR6 and 150W TDP.

InferenceGraphics renderingVirtual workstations

Performance comparison

Visual bars — winner highlighted

NVIDIA L4

Metric

NVIDIA A10

FP16 TFLOPS

242 T
125 T

Bandwidth

300 GB/s
600 GB/s

VRAM

246 GB
246 GB

TDP (lower=better)

72 W
150 W

TFLOPS/watt

3.36
0.83

GB/s per watt

4.17
4.00

Full specification table

SpecNVIDIA L4NVIDIA A10
ArchitectureAda LovelaceAmpere
VRAM24 GB GDDR624 GB GDDR6
VRAM typeGDDR6GDDR6
Memory bandwidth300 GB/s600 GB/s
FP16 TFLOPS242 TFLOPS125 TFLOPS
TDP72 W150 W
TFLOPS/watt3.36 T/W0.83 T/W
GB/s per watt4.17 GB/s/W4.00 GB/s/W
NVLinkNoNo
Release year20232021

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 L4

FP16 TFLOPS242 TFLOPS
TDP72 W
TFLOPS/watt3.361 T/W
GB/s per watt4.17 GB/s/W

NVIDIA A10

FP16 TFLOPS125 TFLOPS
TDP150 W
TFLOPS/watt0.833 T/W
GB/s per watt4.00 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 L4 for:

  • Efficient inference
  • Video transcoding
  • Edge AI

Choose NVIDIA A10 for:

  • Inference
  • Graphics rendering
  • Virtual workstations