Inferentia2 vs Vera Rubin
| Inferentia2 | Vera Rubin | |
|---|---|---|
| Vendor | AWS | NVIDIA |
| Architecture | Custom | Rubin |
| BF16 TFLOPS | 190 | 35,000 ✓ |
| Memory (GB) | 32 | 288 ✓ |
| Bandwidth (GB/s) | 820 | 22,000 ✓ |
| TDP (W) | 200 ✓ | 1,400 |
| Cloud $/hr | $0.55 | — |
| TFLOPS / W | 0.9 | 25.0 ✓ |
FAQ
Inferentia2 vs Vera Rubin: which is faster?
Vera Rubin has higher BF16 throughput (190 vs 35,000 TFLOPS).
Which has more memory?
Vera Rubin carries more on-package memory (32 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.