Beam vs CoreWeave
As of , Beam lists the lower on-demand rate on 5 of the 5 GPUs both offer on-demand. On the priciest shared card both list on-demand, NVIDIA B200: Beam $4.11 vs CoreWeave $8.60 per GPU-hour on-demand — Beam is $4.49 (52%) cheaper.
Cloud GPU rental prices of Beam and CoreWeave, 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.
On-demand, per GPU-hour
| GPU | Beam | CoreWeave | Difference | Cheaper |
|---|---|---|---|---|
| NVIDIA B200 | $4.11 from captured Deploy ↗ | $8.60 captured Deploy ↗ | $4.49 (52%) | Beam |
| NVIDIA H200 | $2.09 from captured Deploy ↗ | $6.30 captured Deploy ↗ | $4.21 (67%) | Beam |
| NVIDIA A100 80GB | $1.36 from captured Deploy ↗ | $2.70 captured Deploy ↗ | $1.34 (50%) | Beam |
| NVIDIA RTX PRO 6000 Blackwell | $1.09 from captured Deploy ↗ | $2.50 captured Deploy ↗ | $1.41 (56%) | Beam |
| NVIDIA L40S | $0.76 from captured Deploy ↗ | $2.25 captured Deploy ↗ | $1.49 (66%) | Beam |
Spot / interruptible, per GPU-hour
| GPU | Beam | CoreWeave |
|---|---|---|
| NVIDIA B200 | — | $4.26 captured |
| NVIDIA H200 | — | $2.58 captured |
| NVIDIA A100 80GB | — | $1.19 captured |
| NVIDIA RTX PRO 6000 Blackwell | — | $1.20 captured |
| NVIDIA L40S | — | $0.98 captured |
GPU by GPU
- NVIDIA B200: Beam $4.11 vs CoreWeave $8.60 per GPU-hour on-demand — Beam is $4.49 (52%) cheaper.
- NVIDIA H200: Beam $2.09 vs CoreWeave $6.30 per GPU-hour on-demand — Beam is $4.21 (67%) cheaper.
- NVIDIA A100 80GB: Beam $1.36 vs CoreWeave $2.70 per GPU-hour on-demand — Beam is $1.34 (50%) cheaper.
- NVIDIA RTX PRO 6000 Blackwell: Beam $1.09 vs CoreWeave $2.50 per GPU-hour on-demand — Beam is $1.41 (56%) cheaper.
- NVIDIA L40S: Beam $0.76 vs CoreWeave $2.25 per GPU-hour on-demand — Beam is $1.49 (66%) 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: