Head to head · updated 2026-09-10
H100-SXM vs H200-SXM: price and deliverability
H200-SXM currently rents at 1.6× the median price of H100-SXM ($6.20 vs $3.99 per GPU-hour on demand). Whether that premium is worth it depends on memory, fabric, and what you are actually running — not the sticker.
| H100-SXM | H200-SXM | |
|---|---|---|
| Lowest on-demand $/GPU-hr | $3.20 | $3.99 |
| Median on-demand | $3.99 | $6.20 |
| Sellers tracked | 12 | 9 |
| Grade A/B offerings | 16 | 13 |
| Max VRAM per GPU (tracked SKUs) | 80 GB | 141 GB |
Memory is the first gate: H200-SXM carries 141 GB per GPU vs 80 GB. If your model + KV cache does not fit, price is irrelevant — the smaller card is not a substitute at any discount (ASSAY-1 §5).
Same-name capacity is not one good: within each column, prices spread by multiples depending on fabric. Check the per-GPU pages for every listing with its published interconnect, and the substitutability rules for when one can genuinely replace the other.
These ratios move weekly
The Delivered Compute Report tracks the spread — free, every row sourced.
Subscribe free