Compute Comparison
vs
All comparisons →

Ada Lovelace

NVIDIA L40

wins

4

The NVIDIA L40 is the predecessor to the L40S, offering 48GB GDDR6 and Ada Lovelace architecture.

Professional inferenceRenderingVisualization AI

Ada Lovelace

NVIDIA RTX 5000 Ada 32GB

wins

1

The RTX 5000 Ada is a professional Ada Lovelace GPU with 32 GB GDDR6 and 130.6 TFLOPS FP16.

Mid-range professional inference30B model servingCost-efficient enterprise AI

Performance comparison

Visual bars — winner highlighted

NVIDIA L40

Metric

NVIDIA RTX 5000 Ada 32GB

FP16 Tensor TFLOPS (dense)

181 T
131 T

Bandwidth

864 GB/s
576 GB/s

VRAM

48 GB
32 GB

TDP (lower=better)

300 W
250 W

TFLOPS/watt

0.60
0.52

GB/s per watt

2.88
2.30

Full specification table

SpecNVIDIA L40NVIDIA RTX 5000 Ada 32GB
ArchitectureAda LovelaceAda Lovelace
VRAM48 GB GDDR632 GB GDDR6
VRAM typeGDDR6GDDR6
Memory bandwidth864 GB/s576 GB/s
FP16 Tensor TFLOPS (dense)181 TFLOPS130.6 TFLOPS
TDP300 W250 W
TFLOPS/watt0.60 T/W0.52 T/W
GB/s per watt2.88 GB/s/W2.30 GB/s/W
NVLinkNoNo
Release year20222023

Live cloud pricing

On-demand pricing across providers — updated in real time

Per hourPer GB VRAM/hr

NVIDIA L40

QuantaCloud
$0.94/hr

NVIDIA RTX 5000 Ada 32GB

No live pricing data available.

Power efficiency analysis

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

NVIDIA L40

FP16 TFLOPS181 TFLOPS
TDP300 W
TFLOPS/watt0.603 T/W
GB/s per watt2.88 GB/s/W

NVIDIA RTX 5000 Ada 32GB

FP16 TFLOPS130.6 TFLOPS
TDP250 W
TFLOPS/watt0.522 T/W
GB/s per watt2.30 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.

Performance-per-dollar

TFLOPS/$ and GB/s/$ at current best cloud prices

NVIDIA L40

Best cloud price$0.94/hr
TFLOPS per $193.1 T/$
GB/s per $922 GB/s/$

NVIDIA RTX 5000 Ada 32GB

Best cloud priceN/A
TFLOPS per $—
GB/s per $—

TFLOPS/$ = FP16 TFLOPS ÷ best on-demand $/hr. Higher is better — more compute per dollar. GB/s/$ = memory bandwidth ÷ best on-demand $/hr. For memory-bound inference workloads (most LLM serving), GB/s/$ is often a better predictor of cost-efficiency than TFLOPS/$. Prices are live from the Compute Comparison API and update in real time.

Workload Cost Comparison

Total job cost = best on-demand $/hr × GPU count × hours

WorkloadScaleNVIDIA L40NVIDIA RTX 5000 Ada 32GB
7B LLM fine-tune8× GPU, 24 h$180—
70B LLM fine-tune64× GPU, 72 h$4.3k—
Inference server1× GPU, 730 h$684—
Image gen batch4× GPU, 8 h$30—

Based on current best on-demand prices. Spot pricing typically 40–70% lower. Actual cost depends on provider, region, and availability.

When to choose each GPU

Choose NVIDIA L40 for:

  • Professional inference
  • Rendering
  • Visualization AI

Weaknesses / Limitations

  • Lower FP16 TFLOPS (181) than L40S (362) at the same VRAM and bandwidth
  • Superseded by L40S — less available on new deployments
  • GDDR6 bandwidth lower than HBM-based data center GPUs

Choose NVIDIA RTX 5000 Ada 32GB for:

  • Mid-range professional inference
  • 30B model serving
  • Cost-efficient enterprise AI

Weaknesses / Limitations

  • GDDR6 memory bandwidth far below HBM alternatives
  • No NVLink — single-card VRAM ceiling
  • Professional but older architecture — no FP8 support