GPU.ai vs Thunder Compute
As of , GPU.ai and Thunder Compute each list the lower on-demand rate on 2 of the 4 GPUs both offer on-demand. On the priciest shared card both list on-demand, NVIDIA H100 SXM: GPU.ai $3.49 vs Thunder Compute $3.20 per GPU-hour on-demand — Thunder Compute is $0.29 (8%) cheaper.
On-demand, per GPU-hour
| GPU | GPU.ai | Thunder Compute | Difference | Cheaper |
|---|---|---|---|---|
| NVIDIA H100 SXM median $3.49 · 31 providers · lowest Cudo Compute $1.79 GPU only | $3.49 from captured Deploy ↗ | $3.20 captured Deploy ↗ | $0.29 (8%) | Thunder Compute |
| NVIDIA A100 80GB median $1.80 · 46 providers · lowest Deep Infra $0.89 | $1.07 from captured Deploy ↗ | $1.09 captured Deploy ↗ | $0.02 (2%) | GPU.ai |
| NVIDIA L40 median $0.90 · 8 providers · lowest GPU.ai $0.78 from | $0.78 from captured Deploy ↗ | $0.79 captured Deploy ↗ | $0.01 (1%) | GPU.ai |
| NVIDIA RTX A6000 median $0.61 · 16 providers · lowest Thunder Compute $0.35 | $0.40 from captured Deploy ↗ | $0.35 captured Deploy ↗ | $0.05 (13%) | Thunder Compute |
How this is compared
Cloud GPU rental prices of GPU.ai and Thunder Compute, compared per GPU-hour on the GPUs both list. Where a provider has several tiers or regions, its cheapest on-demand rate is used and labelled. Community capacity (RunPod Community Cloud) and marketplace floors are shown only where a provider has nothing else, and never decide which is cheaper.
GPU by GPU
- NVIDIA H100 SXM: GPU.ai $3.49 vs Thunder Compute $3.20 per GPU-hour on-demand — Thunder Compute is $0.29 (8%) cheaper.
- NVIDIA A100 80GB: GPU.ai $1.07 vs Thunder Compute $1.09 per GPU-hour on-demand — GPU.ai is $0.02 (2%) cheaper.
- NVIDIA L40: GPU.ai $0.78 vs Thunder Compute $0.79 per GPU-hour on-demand — GPU.ai is $0.01 (1%) cheaper.
- NVIDIA RTX A6000: GPU.ai $0.40 vs Thunder Compute $0.35 per GPU-hour on-demand — Thunder Compute is $0.05 (13%) cheaper.
Not compared: GPUs only one of the two lists, reserved/committed tiers, and anything either provider prices only on request. See each provider's page for its full list.
Data updated: