Compute Comparison
vs
All comparisons →

Ada Lovelace

NVIDIA RTX 4090

wins

1

The RTX 4090 is the most cost-effective consumer GPU for AI workloads. At $0.40–$0.80/hr on GPU clouds, it's ideal for fine-tuning and small-model inference.

Fine-tuningSmall model inferenceCost-effective compute

Ada Lovelace

NVIDIA L40S

wins

4

The L40S is NVIDIA's inference-optimized Ada Lovelace GPU. 48GB GDDR6 and 362 TFLOPS FP16 make it a strong choice for production inference at lower cost than H100.

InferenceVideo AIMulti-modal workloads

Performance comparison

Visual bars — winner highlighted

NVIDIA RTX 4090

Metric

NVIDIA L40S

FP16 TFLOPS

165 T
362 T

Bandwidth

1.0k GB/s
864 GB/s

VRAM

246 GB
486 GB

TDP (lower=better)

450 W
350 W

TFLOPS/watt

0.37
1.03

GB/s per watt

2.24
2.47

Full specification table

SpecNVIDIA RTX 4090NVIDIA L40S
ArchitectureAda LovelaceAda Lovelace
VRAM24 GB GDDR6X48 GB GDDR6
VRAM typeGDDR6XGDDR6
Memory bandwidth1,008 GB/s864 GB/s
FP16 TFLOPS165 TFLOPS362 TFLOPS
TDP450 W350 W
TFLOPS/watt0.37 T/W1.03 T/W
GB/s per watt2.24 GB/s/W2.47 GB/s/W
NVLinkNoNo
Release year20222023

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 4090

FP16 TFLOPS165 TFLOPS
TDP450 W
TFLOPS/watt0.367 T/W
GB/s per watt2.24 GB/s/W

NVIDIA L40S

FP16 TFLOPS362 TFLOPS
TDP350 W
TFLOPS/watt1.034 T/W
GB/s per watt2.47 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 4090 for:

  • Fine-tuning
  • Small model inference
  • Cost-effective compute

Choose NVIDIA L40S for:

  • Inference
  • Video AI
  • Multi-modal workloads