The NVIDIA H200 is an enterprise-class GPU built on the Hopper architecture, featuring 141 GB of memory with 4,800 GB/s of memory bandwidth and a 700 W TDP. It delivers 67 TFLOPS of FP32 performance and 989 TFLOPS of FP16 tensor performance. Currently available from 7 providers starting at $1.32/GPU/hour, with a market median of $4.00/GPU/hour across 20 configurations.
Cheapest
$1.320/hr
Vast.ai
Median
$3.995/hr
across 20 offerings
Most expensive
$6.180/hr
Vast.ai
90-day trend
+6.5%
median rate
Hardware specifications
Same across all providers
Memory
141 GB
HBM3
FP16 performance
989 TFLOPS
peak
TDP
700 W
Memory bandwidth
4,800 GB/s
Architecture
Hopper
FP32 performance
67 TFLOPS
Brand
NVIDIA
Series
H200
Market price history
Median price across all providers
9 offerings from 7 providers
Sorted by price ascending. Compare configurations side-by-side.
| Provider | Count | vCPU | RAM | Region | Per GPU hour | Total/hr | Action |
|---|---|---|---|---|---|---|---|
| ×1 | 32 | 62 GB |    +1 | From$1.320 +0.5% 30d | From$1.32 | Launch | |
| ×1 | 8 | 141 GB | -- | $2.490 | From$2.49 | Launch | |
| ×1 | 24 | 240 GB | -- | $3.440 0.0% 30d | From$3.44 | ||
 OblivusReferral link | ×1 | — | — | | $3.990 | $3.99 | Launch |
| ×8 | 176 | 1,800 GB |  | $3.990 +14.0% 30d | $31.92 | ||
| ×1 | 44 | 182 GB | -- | $4.000 | From$4.00 | Launch | |
 RunpodReferral link | ×1 | 12 | 188 GB |  +1 | $4.390 | $4.39 | Launch |
 Sesterce CloudReferral link | ×2 | 88 | 370 GB |  | From$5.077 +7.1% 30d | From$10.15 | Launch |
| ×2 | 176 | 740 GB | | $5.080 +34.4% 30d | $10.16 | Launch |
Similar GPUs
See more GPUs like this in
Want more from this page?
See incorrect data?
Frequently asked questions
How much does the H200 cost per hour?
The H200 starts at $1.32 per GPU-hour, with a median of $4.00 across 20 available configurations. Actual pricing depends on the provider, region, and whether you rent on-demand, spot or reserved.
Which cloud providers offer the H200?
10 providers currently list the H200: Vast.ai, AceCloud, Hyperstack, Runcrate, Oblivus, Verda and 4 more. Compare their hourly pricing, regions and machine specs in the table above.
How much VRAM does the H200 have?
The H200 comes with 141 GB of VRAM, which determines the largest models and batch sizes it can hold in memory for training and inference.
Where is the H200 cheapest?
The lowest on-demand price we currently track for the H200 is $1.32 per GPU-hour at Vast.ai. Because providers update pricing and availability frequently, check the live table above before you launch.