Compute Comparison
vs
All comparisons →

Ampere

NVIDIA A40

wins

1

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

Large model inferenceProfessional renderingMulti-tenant AI

Blackwell Ultra

NVIDIA GB300 NVL72

wins

4

The GB300 NVL72 is NVIDIA's Blackwell Ultra rack-scale system — 72 GPUs in a single rack delivering 7,500 TFLOPS FP16 per GPU.

Next-gen frontier trainingTrillion-param inferenceRack-scale AI

Performance comparison

Visual bars — winner highlighted

NVIDIA A40

Metric

NVIDIA GB300 NVL72

FP16 Tensor TFLOPS (dense)

150 T
7.5k T

Bandwidth

696 GB/s
16.0k GB/s

VRAM

48 GB
288 GB

TDP (lower=better)

300 W
1.2k W

TFLOPS/watt

0.50
6.25

GB/s per watt

2.32
13.33

Full specification table

SpecNVIDIA A40NVIDIA GB300 NVL72
ArchitectureAmpereBlackwell Ultra
VRAM48 GB GDDR6288 GB HBM3e (per GPU)
VRAM typeGDDR6HBM3e (per GPU)
Memory bandwidth696 GB/s16,000 GB/s
FP16 Tensor TFLOPS (dense)149.7 TFLOPS7,500 TFLOPS
TDP300 W1,200 W
TFLOPS/watt0.50 T/W6.25 T/W
GB/s per watt2.32 GB/s/W13.33 GB/s/W
NVLinkNoYes
Release year20202026

Live cloud pricing

On-demand pricing across providers — updated in real time

Per hourPer GB VRAM/hr
Loading prices…

Power efficiency analysis

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

NVIDIA A40

FP16 TFLOPS149.7 TFLOPS
TDP300 W
TFLOPS/watt0.499 T/W
GB/s per watt2.32 GB/s/W

NVIDIA GB300 NVL72

FP16 TFLOPS7,500 TFLOPS
TDP1,200 W
TFLOPS/watt6.250 T/W
GB/s per watt13.33 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 A40 for:

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

Weaknesses / Limitations

  • Older Ampere architecture — lower FP16 TFLOPS than L40S at same VRAM
  • GDDR6 bandwidth (696 GB/s) lower than HBM-based GPUs
  • No NVLink on A40

Choose NVIDIA GB300 NVL72 for:

  • Next-gen frontier training
  • Trillion-param inference
  • Rack-scale AI

Weaknesses / Limitations

  • Rack-scale system — not a single GPU, requires full NVL72 rack deployment
  • Extremely high cost — rack-scale pricing only
  • Very limited availability, primarily hyperscaler deployments