H100 · B200
Read 12 November 2026
H100 against B200: price and memory
- B200 lowest, against H100
- +76%
- $3.50 against $1.99 per GPU-hour
- H100 median
- $3.90/GPU-hr
- 19 providers
- B200 median
- $8.34/GPU-hr
- 10 providers
- List both
- 10
- providers
Basis: each GPU's lowest price per provider on current on-demand rows the provider sets. The gap is (B200 price minus H100 price) divided by the H100 price. Marketplace figures and announced prices are left out. USD per GPU-hour, before tax.
Side by side
derived columns are labelled| Measure | H100 | B200 | B200 against H100 |
|---|---|---|---|
| Lowest $/GPU-hr | $1.99 | $3.50 | +76% |
| Median $/GPU-hr | $3.90 | $8.34 | +114% |
| Highest $/GPU-hr | $11.06 | $14.24 | +29% |
| Providers with a price | 19 | 10 | 10 list both |
| VRAM, GB (catalogue) | 80 | 180 | +125% |
| Lowest $ per GB-hour (derived) | $0.0249 | $0.0194 | -22% |
| Median $ per GB-hour (derived) | $0.0488 | $0.0464 | -5% |
Price per GB-hour = price per GPU-hour divided by the catalogue VRAM in GB: $1.99 / 80 = $0.0249 for the H100 and $3.50 / 180 = $0.0194 for the B200. Memory is only one of the things a GPU rental buys; this column says nothing about speed.
Memory as the maker states it
catalogue VRAM is used for the price per GB| GPU | Variant | Memory GB | Memory type | Bandwidth TB/s | Source |
|---|---|---|---|---|---|
| H100 | SXM | 80 | not stated | 3.35 | maker page |
| H100 | NVL | 94 | HBM3 | 3.9 | maker page |
| B200 | SXM | 192 | HBM3E | 8 | maker page |
Figures are copied from each maker page with the variant it names; where a page does not state a figure the cell says not stated. The price per GB above uses the catalogue VRAM (80 GB and 180 GB) so that every GPU in the data has one comparable number.
Providers that list both
10 providersAt 0 of 10 of these providers the B200 is the lower price. The gap from the H100 runs from +40% to +176%, with a median of +94% across the providers.
H100 or B200: common questions
Is the B200 lower priced than the H100?
Compare the lowest listings first: the B200 is $3.50 against $1.99 for the H100, 76% above the H100. On the median provider low it is $8.34 against $3.90, 114% above the H100.
Which costs less per GB of memory?
At the lowest listing the B200 is $0.0194 per GB-hour and the H100 is $0.0249, so the B200 is 22% below the H100 on that measure. At the median the figures are $0.0464 and $0.0488. This uses the catalogue VRAM of 180 GB and 80 GB.
How much memory does the B200 have compared with the H100?
180 GB against 80 GB in the catalogue, which is 125% above the H100 (2.25 times). The maker table above lists the stated figure for each variant.
Which providers list both?
10 providers: Vultr, Hyperstack, Massed Compute, RunPod, Together AI, Nebius, Verda, and CoreWeave and 2 more.
Does the price gap hold at the same provider?
At 0 of 10 providers the B200 is lower than that provider's own H100 price. The gaps run from +40% to +176%, so the market-wide gap of +114% between medians is not the gap you will see at every provider.
Is Runcrate included in these numbers?
Yes, where it has a live price: H100 $3.58 per GPU-hour. It is sorted by price like every other row and carries no pinned position.
Disclosure
H100 Rate Card is published by the Runcrate team, who also run Runcrate, an AI cloud. Runcrate is listed in the table for the H100 and B200, marked "ours, live". It is sorted by price like every other row, with no pinned position. Other providers' prices are read from their own public pricing pages. We have no paid placement.