Back to all offerings
NVIDIA H100
80 GB VRAM989 FP16 TFLOPS700 W TDP11 providers49 offerings13 regions
The NVIDIA H100 is an enterprise-class GPU built on the Hopper architecture, featuring 80 GB of memory with 3,350 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 11 providers starting at $1.33/GPU/hour, with a market median of $2.94/GPU/hour across 49 configurations.
Cheapest
$1.330/hr
Vast.ai
Median
$2.940/hr
across 49 offerings
Most expensive
$6.589/hr
Sesterce Cloud
90-day trend
+9.3%
median rate
Hardware specifications
Same across all providers
Memory
80 GB
HBM3
FP16 performance
989 TFLOPS
peak
TDP
700 W
Memory bandwidth
3,350 GB/s
Architecture
Hopper
FP32 performance
67 TFLOPS
Brand
NVIDIA
Series
H100
Market price history
Median price across all providers
13 offerings from 11 providers
Sorted by price ascending. Compare configurations side-by-side.
| Provider | Count | vCPU | RAM | Region | Per GPU hour | Total/hr | Action |
|---|---|---|---|---|---|---|---|
| ×8 | 192 | 1,512 GB |    +4 | From$1.330 -14.5% 30d | From$10.64 | Launch | |
 Thunder ComputeReferral link | ×1 | 4 | 32 GB | -- | From$2.190 -2.0% 30d | From$2.19 | Launch |
| ×1 | 10 | 80 GB | -- | $2.290 | From$2.29 | Launch | |
 RunpodReferral link | ×1 | 16 | 176 GB |   +4 | From$2.890 +1.7% 30d | From$2.89 | Launch |
| ×8 | 160 | 1,920 GB | -- | From$2.990 0.0% 30d | From$23.92 | ||
| ×8 | 112 | 1,880 GB | | $3.200 +31.6% 30d | $25.60 | ||
| ×1 | 30 | 120 GB | -- | $3.250 | From$3.25 | Launch | |
| ×1 | 24 | 240 GB |   | From$3.277 -0.9% 30d | From$3.28 | Launch | |
| ×1 | 26 | 200 GB | | From$3.290 -23.3% 30d | From$3.29 | Launch | |
 Novita AIReferral link | ×1 | 16 | 128 GB |  | $3.390 +30.9% 30d | $3.39 | Launch |
 OblivusReferral link | ×1 | 28 | 180 GB | | From$3.500 +60.7% 30d | From$3.50 | Launch |
| ×1 | 24 | 240 GB |  +1 | $3.630 +73.7% 30d | From$3.63 | Launch | |
 Sesterce CloudReferral link | ×2 | 48 | 480 GB | | From$3.630 +40.8% 30d | From$7.26 | Launch |
Similar GPUs
See more GPUs like this in
Want more from this page?
See incorrect data?
Frequently asked questions
How much does the H100 cost per hour?
The H100 starts at $1.33 per GPU-hour, with a median of $2.94 across 49 available configurations. Actual pricing depends on the provider, region, and whether you rent on-demand, spot or reserved.
Which cloud providers offer the H100?
14 providers currently list the H100: Scaleway, Novita AI, Lambda, Theta EdgeCloud, AceCloud, DigitalOcean and 8 more. Compare their hourly pricing, regions and machine specs in the table above.
How much VRAM does the H100 have?
The H100 comes with 80 GB of VRAM, which determines the largest models and batch sizes it can hold in memory for training and inference.
Where is the H100 cheapest?
The lowest on-demand price we currently track for the H100 is $1.33 per GPU-hour at Vast.ai. Because providers update pricing and availability frequently, check the live table above before you launch.