Compute Comparison
vs
All comparisons →

Ampere

NVIDIA RTX A6000

wins

3

The RTX A6000 is NVIDIA's Ampere professional workstation GPU with 48GB GDDR6 and NVLink support.

Professional workstation AILarge model inferenceRendering

Ampere

NVIDIA A40

wins

0

The NVIDIA A40 offers 48GB GDDR6 at a lower price point than the A100.

Large model inferenceProfessional renderingMulti-tenant AI

Performance comparison

Visual bars — winner highlighted

NVIDIA RTX A6000

Metric

NVIDIA A40

FP16 TFLOPS

154 T
150 T

Bandwidth

768 GB/s
696 GB/s

VRAM

486 GB
486 GB

TDP (lower=better)

300 W
300 W

TFLOPS/watt

0.51
0.50

GB/s per watt

2.56
2.32

Full specification table

SpecNVIDIA RTX A6000NVIDIA A40
ArchitectureAmpereAmpere
VRAM48 GB GDDR648 GB GDDR6
VRAM typeGDDR6GDDR6
Memory bandwidth768 GB/s696 GB/s
FP16 TFLOPS154 TFLOPS150 TFLOPS
TDP300 W300 W
TFLOPS/watt0.51 T/W0.50 T/W
GB/s per watt2.56 GB/s/W2.32 GB/s/W
NVLinkYesNo
Release year20202020

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 RTX A6000

FP16 TFLOPS154 TFLOPS
TDP300 W
TFLOPS/watt0.513 T/W
GB/s per watt2.56 GB/s/W

NVIDIA A40

FP16 TFLOPS150 TFLOPS
TDP300 W
TFLOPS/watt0.500 T/W
GB/s per watt2.32 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 RTX A6000 for:

  • Professional workstation AI
  • Large model inference
  • Rendering

Choose NVIDIA A40 for:

  • Large model inference
  • Professional rendering
  • Multi-tenant AI