Aquanode DocsGPUs & Pricing
Pricing
GPU pricing models and on-demand rates across all Aquanode providers
Pricing varies by provider and GPU model. Aquanode surfaces the best available option based on your requirements.
On-demand pricing varies
Prices shown are indicative starting rates. Actual on-demand pricing fluctuates based on availability, demand, and provider conditions. Check the Aquanode Marketplace for live rates.
Pricing by Provider
| Provider | GPU Model(s) | On-Demand Price/hr | Notes |
|---|---|---|---|
| Voltage Park | HGX H100 | from $1.99/hr | Long-term reservations available (6+ months) |
| DataCrunch | H100, H200, B200, A100 | from $1.50/hr | Fixed + dynamic pricing, contract options |
| Hot Aisle | AMD MI300x, MI355x | from $1.95/hr | On-demand AMD GPUs |
| Akash Network | H100, A100, RTX series | varies (reverse auction) | Decentralized marketplace, often cheapest |
| Hyperstack | H100, A100, L40, A6000 | from $1.20/hr | Cost-effective, Canada-based |
| Vast.ai | A100, RTX 4090, L40, V100 | from $0.50/hr | Decentralized, community + datacenter sourced |
| Vultr | A16, A40, A100 | from $0.20/hr | Virtualized, low-VRAM options |
| Massed Compute | A6000, L40, A100, H100 | from $0.80/hr | Small-scale enterprise, US-based |
| Cudo Compute | L40, A100, H100 | from $1.50/hr | Cost-effective, Norway-based |
Pricing Models
- Voltage Park — On-demand with long-term reservation discounts for large-scale workloads
- DataCrunch — Fixed base rate + dynamic pricing on storage configured
- Hot Aisle — On-demand AMD Fixed GPU pricing
- Akash, Vast.ai — Decentralized auction marketplaces; prices fluctuate with supply and demand, typically the lowest cost options
- Hyperstack, Massed Compute, Cudo Compute — On-demand with competitive fixed rates
- Vultr — On-demand, virtualized infrastructure
For live pricing, see the Aquanode Marketplace.