The NVIDIA H200 (Hopper, 141 GB HBM3e) rents from $2.81/hr, averaged across 10 cloud providers, updated 2026-06-30. On-demand pricing has risen 7% over the past 38 days.
An H100 with ~1.8× the memory and bandwidth — same core compute, but fits larger models and serves long-context inference more efficiently. Hopper · 141 GB HBM3e · 4.8 TB/s memory bandwidth · launched 2024.
▮ on-demand /hr · ▮ spot /hr
| Date | On-demand | Spot |
|---|---|---|
| 2026-06-30 | $4.17 | $2.81 |
| 2026-06-29 | $4.13 | $2.84 |
| 2026-06-28 | $4.10 | $2.70 |
| 2026-06-27 | $4.13 | $2.81 |
| 2026-06-26 | $4.17 | $2.66 |
| 2026-06-24 | $4.11 | $2.99 |
| 2026-06-23 | $4.18 | $3.28 |
| 2026-06-19 | $4.04 | $2.81 |
| 2026-06-18 | $3.95 | $3.02 |
| 2026-06-17 | $4.01 | $2.96 |
| 2026-06-15 | $4.02 | $3.08 |
| 2026-06-14 | $4.02 | $2.86 |
| 2026-06-13 | $4.02 | $2.93 |
| 2026-06-12 | $4.09 | $3.02 |
| GPU | On-demand /hr | Spot /hr | Memory | |
|---|---|---|---|---|
| H200 | $4.17 | $2.81 | 141 GB HBM3e | |
| B200 | $6.77 | $3.51 | 192 GB HBM3e | H200 vs B200 → |
| H100 | $3.55 | $1.59 | 80 GB HBM3 | H200 vs H100 → |
| MI355X | $8.60 | $— | 288 GB HBM3e | H200 vs MI355X → |
| MI325X | $3.26 | $— | 256 GB HBM3e | H200 vs MI325X → |
| MI300X | $3.10 | $1.20 | 192 GB HBM3 | H200 vs MI300X → |
The H200 averages $4.17/hr on-demand, $2.81/hr spot, and $3.42/hr reserved across 10 cloud providers (updated 2026-06-30).
The lowest tier is spot at about $2.81/hr on average; reserved (1–3 year commitment) runs ~$3.42/hr. Compare live per-provider rates on the Bargo GPU price tracker.
At on-demand $4.17/hr, one H200 runs about $3,044/month (730 hours); reserved pricing lowers that to roughly $2,497/month.
On-demand, the H200 is $4.17/hr vs the H100 at $3.55/hr — pricier per hour. See the full H200 vs H100 comparison for specs and cost-per-job context.
Spot ($2.81/hr) is interruptible spare capacity at a discount to on-demand ($4.17/hr); use spot for checkpointed or fault-tolerant jobs and on-demand or reserved for long, uninterrupted runs.
Cross-provider averages from public cloud GPU listings. Informational only.