L40S vs Vera Rubin
| L40S | Vera Rubin | |
|---|---|---|
| Vendor | NVIDIA | NVIDIA |
| Architecture | Ada Lovelace | Rubin |
| BF16 TFLOPS | 362 | 35,000 ✓ |
| Memory (GB) | 48 | 288 ✓ |
| Bandwidth (GB/s) | — | 22,000 |
| TDP (W) | 350 ✓ | 1,400 |
| Cloud $/hr | — | — |
| TFLOPS / W | 1.0 | 25.0 ✓ |
FAQ
L40S vs Vera Rubin: which is faster?
Vera Rubin has higher BF16 throughput (362 vs 35,000 TFLOPS).
Which has more memory?
Vera Rubin carries more on-package memory (48 GB vs 288 GB).
Which is cheaper to rent?
Cloud rates aren't listed for both chips — see each chip's page.
Cost either chip in a fleet
Size a deployment or see the full priced rack BOM these accelerators sit in.
Specs + modeled rates maintained by Mintok's catalog watchers from public sources. Not investment advice. Methodology.