NVIDIA L40S 48GB vs NVIDIA RTX 6000 Ada 48GB
L40S 48GB has about the same memory than RTX 6000 Ada 48GB (48 GB vs 48 GB). Its memory bandwidth is 10% less (0.864 vs 0.96 TB/s). In the cloud, the median on-demand L40S 48GB price is $1.50 per GPU-hour against $0.79 for the RTX 6000 Ada 48GB. Per gigabyte of GPU memory, that is $0.031 vs $0.016 per hour.
Specifications
Manufacturer figures as recorded by Kovara. Higher memory and bandwidth are highlighted; neither is a benchmark of your workload.
| L40S 48GB | RTX 6000 Ada 48GB | |
|---|---|---|
| Manufacturer | NVIDIA | NVIDIA |
| Architecture | Ada Lovelace | Ada Lovelace |
| Memory | 48 GB GDDR6 | 48 GB GDDR6 |
| Memory bandwidth | 0.864 TB/s | 0.96 TB/s |
| Power (TDP) | 350 W | 300 W |
| Form factor | PCIe dual-slot | PCIe dual-slot |
Cloud prices today
Public offers Kovara tracks, normalised to US dollars per GPU-hour. Prices use on-demand terms where available; spot and reserved offers are on each GPU’s price pages. Lower is highlighted.
| L40S 48GB | RTX 6000 Ada 48GB | |
|---|---|---|
| Lowest price | $0.97 | $0.79 |
| Median price | $1.50 | $0.79 |
| Highest price | $2.25 | $1.64 |
| Median per GB of memory | $0.031 | $0.016 |
| Public offers (all terms) | 38 | 9 |
| Providers | 9 | 4 |
Where to rent each
Lowest tracked price per provider. Open a provider to see every offer, region and term.
Questions
Which is cheaper to rent, the L40S 48GB or the RTX 6000 Ada 48GB?
Right now the cheapest on-demand L40S 48GB Kovara tracks is $0.97 per GPU-hour and the cheapest RTX 6000 Ada 48GB is $0.79. Medians are $1.50 and $0.79. The cheaper GPU per hour is not always cheaper per job: a faster or larger GPU can finish sooner or need fewer units.
Which has more memory?
Both have 48 GB of GPU memory.
Are these benchmark results?
No. This page compares manufacturer specifications and public prices. Real performance depends on the model, precision, software and interconnect. Kova can help you estimate cost for a specific workload.