EmpirioLabs vs Servercore
As of , EmpirioLabs lists the lower on-demand rate on 4 of the 4 GPUs both offer on-demand. On the priciest shared card both list on-demand, NVIDIA H200: EmpirioLabs $8.00 vs Servercore $8.83 per GPU-hour on-demand — EmpirioLabs is $0.83 (9%) cheaper.
On-demand, per GPU-hour
| GPU |  EmpirioLabs |  Servercore | Difference | Cheaper |
|---|---|---|---|---|
| NVIDIA H200 median $4.56 · 56 providers · lowest Dataoorts $2.29 | $8.00 captured Deploy ↗ | $8.83 captured Deploy ↗ | $0.83 (9%) | EmpirioLabs |
| NVIDIA RTX PRO 6000 Blackwell median $2.20 · 42 providers · lowest Hostinger $0.65 from | $4.00 captured Deploy ↗ | $4.21 captured Deploy ↗ | $0.21 (5%) | EmpirioLabs |
| NVIDIA A100 80GB median $1.99 · 63 providers · lowest Yotta Labs $0.92 | $1.70 captured Deploy ↗ | $6.82 captured Deploy ↗ | $5.12 (75%) | EmpirioLabs |
| NVIDIA RTX A5000 median $0.58 · 10 providers · lowest GPU.ai $0.27 from | $0.70 captured Deploy ↗ | $1.12 captured Deploy ↗ | $0.42 (38%) | EmpirioLabs |
How this is compared
Cloud GPU rental prices of EmpirioLabs and Servercore, 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 H200: EmpirioLabs $8.00 vs Servercore $8.83 per GPU-hour on-demand — EmpirioLabs is $0.83 (9%) cheaper.
- NVIDIA RTX PRO 6000 Blackwell: EmpirioLabs $4.00 vs Servercore $4.21 per GPU-hour on-demand — EmpirioLabs is $0.21 (5%) cheaper.
- NVIDIA A100 80GB: EmpirioLabs $1.70 vs Servercore $6.82 per GPU-hour on-demand — EmpirioLabs is $5.12 (75%) cheaper.
- NVIDIA RTX A5000: EmpirioLabs $0.70 vs Servercore $1.12 per GPU-hour on-demand — EmpirioLabs is $0.42 (38%) 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: