H200 vs MI300X
Datasheet specs side by side, plus what each card actually costs to rent today across every provider we track.
H200
MI300X
Spec sheet
| H200 | MI300X | |
|---|---|---|
| Architecture | Hopper | CDNA 3 |
| VRAM (max variant) | 143 GB | 192 GB |
| FP16 TFLOPS (dense) | 989 | 1307 |
| Memory bandwidth | 4800 GB/s | 5300 GB/s |
| TDP | 700 W | 750 W |
| Interconnect | NVLink 900GB/s | Infinity Fabric |
| Launched | 2023 | 2023 |
Computed price-performance at today's cheapest on-demand rates: H200 $2.63 per PFLOP-hour vs MI300X $1.83. Formula: cheapest $/hr ÷ dense FP16 PFLOPS (1 PFLOP = 1000 TFLOPS). Specs from vendor datasheets (source, source).
What they cost to rent (2026-09-01)
| H200 | MI300X | |
|---|---|---|
| Cheapest on-demand $/hr | $2.60 | $2.39 |
| Cheapest spot/community $/hr | $2.51 | – |
| Cheapest source | GMI Cloud | RunPod |
| Providers offering it | 22 | 4 |
Cheapest verified on-demand rate per side from today's snapshot; highlighted = cheaper. Spot/marketplace tiers shown on the individual pages.
FAQ
MI300X is cheaper: the H200 rents from $2.60 per GPU-hour and the MI300X from $2.39 on-demand, verified 2026-09-01.
The H200 has 143 GB and the MI300X has 192 GB. The MI300X has more.
Prices render from the live dataset each build. More: the screener · price movers · methodology.