GPU prices, per GPU-hour
What it costs to rent a GPU by the hour, and what the same card costs to buy outright. Node-priced offerings are divided by their GPU count, so every figure on this site is directly comparable. Each price links to the provider page it was read from.
Cheapest current rate by model
| GPU | VRAM | Cheapest / GPU-hour | Provider | Type | Retail, new |
|---|---|---|---|---|---|
| NVIDIA B200 | 192 GB | $5.98 | RunPod | on-demand | — |
| NVIDIA H200 | 141 GB | $3.59 | RunPod | on-demand | — |
| NVIDIA H100 SXM | 80 GB | $2.69 | RunPod | on-demand | — |
| NVIDIA H100 PCIe | 80 GB | $1.99 | RunPod | on-demand | — |
| NVIDIA A100 80GB | 80 GB | $0.56 | Vast.ai | on-demand | — |
| NVIDIA L40S | 48 GB | $0.53 | Vast.ai | on-demand | — |
| NVIDIA RTX 4090 | 24 GB | $0.30 | Vast.ai | on-demand | — |
| NVIDIA A100 40GB | 40 GB | — | — | — | — |
Providers
| Provider | Regions | Models tracked | Notes |
|---|---|---|---|
| RunPod | — | 7 | — |
| Vast.ai | — | 7 | auction pricing, API only |
| Cudo Compute | — | 0 | — |
| GMI Cloud | — | 0 | — |
| Nebius | — | 0 | — |
| DataCrunch | — | 0 | — |
| Hyperstack | — | 0 | — |
| Jarvislabs | — | 0 | — |
| Lambda | — | 0 | SXM prices are per 8-GPU node - normalize |
| Modal | — | 0 | per-second serverless |
| OVHcloud | — | 0 | — |
| Scaleway | — | 0 | — |
| TensorDock | — | 0 | marketplace |
| Together AI | — | 0 | — |
| Vultr | — | 0 | — |
| CoreWeave | — | 0 | 8-GPU node prices - normalize; GB200/B300 quote-only |
| Crusoe | — | 0 | Blackwell quote-only |
| Latitude.sh | — | 0 | bare metal |
| Massed Compute | — | 0 | — |
| Paperspace | — | 0 | multi-GPU per-node - normalize |
| Salad | — | 0 | consumer GPUs only |
| Voltage Park | — | 0 | — |
| Sesterce | — | 0 | needs headless browser |
Data updated: