NVIDIA L40SvsNVIDIA RTX 4090
NVIDIA L40S vs NVIDIA RTX 4090: Specs, Benchmarks & Cloud Pricing
Full technical comparison of NVIDIA L40S and NVIDIA RTX 4090 — VRAM, memory bandwidth, FP16 TFLOPS, TDP, and live cloud pricing across all providers. Updated July 2026.
Technical Specifications
Spec
NVIDIA L40S
NVIDIA RTX 4090
Architecture
Ada Lovelace
Ada Lovelace
VRAM
48 GB GDDR6
24 GB GDDR6X
Memory bandwidth
864 GB/s
1,008 GB/s
FP16 performance
362 TFLOPS
165 TFLOPS
TDP
350 W
450 W
NVLink support
No
No
Release year
2023
2022
Green = better value for that metric.
NVIDIA L40S
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.
Best For
- Inference
- Video AI
- Multi-modal workloads
NVIDIA RTX 4090
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.
Best For
- Fine-tuning
- Small model inference
- Cost-effective compute