NVIDIA
NVIDIA B200
Blackwell-generation data-center accelerator used in HGX and DGX systems for large-scale AI training and inference.
- Architecture
- Blackwell
- Memory
- 180 GB HBM3e
- Memory bandwidth
- Up to 8 TB/s
- Release
- Announced 2024
Main use cases
Frontier model training and high-throughput inference
Cloud pricing
Full tracker →| Provider | Price / hour | Region | Availability | Updated |
|---|---|---|---|---|
| CoreWeave | Not verified | US East | Limited | 28 Sept 2026, 17:19 |
| Lambda | USD 6.99 | — | 1-GPU on-demand list rate; tax extra; capacity not checked | 7 Oct 2026, 06:10 |
| Nebius | USD 8.5 | — | Published on-demand USD rate per GPU-hour, effective 1 October 2026; taxes extra; capacity not checked | 7 Oct 2026, 06:04 |
| Runpod | USD 6.79 | — | Secure Cloud Pod; published on-demand rate | 29 Sept 2026, 07:24 |
| Verda | USD 6.11 | — | On-demand GPU instance; published rate | 29 Sept 2026, 07:28 |
Newest-generation capacity; availability is usually the limiting factor rather than list price.
NVIDIA B200 in the news
GPU news →No coverage yet.