GPU cloud · 10 GPU models tracked
CoreWeave
$0.78–$10.50per GPU-hour, the cheapest row to the dearest
CoreWeave prices most GPUs as 8-GPU HGX nodes; the rate here is the node price divided by 8. On-demand and spot. GB200 and B300 on-demand are available by quote only and are not listed.
As of , CoreWeave lists 10 GPU models we track, from $0.78 per GPU-hour (NVIDIA L40, spot) to $10.50 (NVIDIA GB200 NVL72, on-demand).
| Website | coreweave.com |
|---|---|
| Pricing page | coreweave.com/pricing |
| Regions and tiers | Europe, North America — as labelled in the rates below |
| Models tracked | 10 |
Price by GPU
As of , per GPU-hour: CoreWeave’s on-demand rate for each card it lists, a cheaper market it also sells, and where the on-demand rate stands among the providers we track.
| GPU | CoreWeave on-demand, per GPU-hour | Against the market (median rate card) |
|---|---|---|
| NVIDIA GB200 NVL72 | $10.50 4-GPU node captured Deploy ↗ | 21% below the median $13.25 · 6 providers 2 of 6 list lower · lowest Bitdeer AI $6.94 from · compare |
| NVIDIA B200 | $8.60 8-GPU node captured spot $4.26 Deploy ↗ | 15% above the median $7.50 · 34 providers 25 of 34 list lower · lowest Beam $4.11 from · compare |
| NVIDIA GH200 | $6.50 captured Deploy ↗ | 103% above the median $3.20 · 4 providers 3 of 4 list lower · lowest Lambda $2.29 · compare |
| NVIDIA H200 | $6.30 8-GPU node captured spot $2.58 Deploy ↗ | 38% above the median $4.56 · 50 providers 39 of 50 list lower · lowest Beam $2.09 from · compare |
| NVIDIA H100 SXM | $6.16 8-GPU node captured spot $2.44 Deploy ↗ | 75% above the median $3.52 · 38 providers 32 of 38 list lower · lowest Cudo Compute $1.79 GPU only · compare |
| NVIDIA B300 | $4.48 spot · no on-demand rate captured Deploy ↗ | no on-demand rate here, so no position against on-demand rate cards |
| NVIDIA A100 80GB | $2.70 8-GPU node captured spot $1.19 Deploy ↗ | 35% above the median $2.00 · 58 providers 42 of 58 list lower · lowest CloudRift $1.05 · compare |
| NVIDIA RTX PRO 6000 Blackwell | $2.50 8-GPU node captured spot $1.20 Deploy ↗ | 13% above the median $2.21 · 38 providers 25 of 38 list lower · lowest Hostinger $0.65 from · compare |
| NVIDIA L40S | $2.25 8-GPU node captured spot $0.98 Deploy ↗ | 61% above the median $1.40 · 38 providers 34 of 38 list lower · lowest CloudRift $0.63 · compare |
| NVIDIA L40 | $1.25 8-GPU node captured spot $0.78 Deploy ↗ | 31% above the median $0.95 · 10 providers 8 of 10 list lower · lowest GPU.ai $0.78 from · compare |
The same, in words
- CoreWeave lists the NVIDIA GB200 NVL72 from $10.50 per GPU-hour on-demand (4-GPU node). The on-demand rate is unchanged since first read, Sep 7, 2026 (7-day change: none). Against the market: 21% below the median rate card ($13.25 across 6 providers); 2 of the 6 providers we track list a lower on-demand rate, the lowest Bitdeer AI at $6.94 (Bitdeer AI’s rates, CoreWeave vs Bitdeer AI).
- CoreWeave lists the NVIDIA B200 from $8.60 per GPU-hour on-demand (8-GPU node), $4.26 spot (North America, 8-GPU node). The on-demand rate is unchanged since first read, Sep 7, 2026 (7-day change: none). Against the market: 15% above the median rate card ($7.50 across 34 providers); 25 of the 34 providers we track list a lower on-demand rate, the lowest Beam at $4.11 (Beam’s rates, CoreWeave vs Beam).
- CoreWeave lists the NVIDIA GH200 from $6.50 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 8, 2026 (7-day change: none). Against the market: 103% above the median rate card ($3.20 across 4 providers); 3 of the 4 providers we track list a lower on-demand rate, the lowest Lambda at $2.29 (Lambda’s rates, CoreWeave vs Lambda).
- CoreWeave lists the NVIDIA H200 from $6.30 per GPU-hour on-demand (8-GPU node), $2.58 spot (Europe, 8-GPU node). The on-demand rate is unchanged since first read, Sep 7, 2026 (7-day change: none). Against the market: 38% above the median rate card ($4.56 across 50 providers); 39 of the 50 providers we track list a lower on-demand rate, the lowest Beam at $2.09 (Beam’s rates, CoreWeave vs Beam).
- CoreWeave lists the NVIDIA H100 SXM from $6.16 per GPU-hour on-demand (8-GPU node), $2.44 spot (Europe, 8-GPU node). The on-demand rate is unchanged since first read, Sep 7, 2026 (7-day change: none). Against the market: 75% above the median rate card ($3.52 across 38 providers); 32 of the 38 providers we track list a lower on-demand rate, the lowest Cudo Compute at $1.79 (Cudo Compute’s rates, CoreWeave vs Cudo Compute).
- CoreWeave lists the NVIDIA B300 from $4.48 spot (North America, 8-GPU node).
- CoreWeave lists the NVIDIA A100 80GB from $2.70 per GPU-hour on-demand (8-GPU node), $1.19 spot (Europe, 8-GPU node). The on-demand rate is unchanged since first read, Sep 7, 2026 (7-day change: none). Against the market: 35% above the median rate card ($2.00 across 58 providers); 42 of the 58 providers we track list a lower on-demand rate, the lowest CloudRift at $1.05 (CloudRift’s rates, CoreWeave vs CloudRift).
- CoreWeave lists the NVIDIA RTX PRO 6000 Blackwell from $2.50 per GPU-hour on-demand (8-GPU node), $1.20 spot (8-GPU node). The on-demand rate is unchanged since first read, Sep 7, 2026 (7-day change: none). Against the market: 13% above the median rate card ($2.21 across 38 providers); 25 of the 38 providers we track list a lower on-demand rate, the lowest Hostinger at $0.65 (Hostinger’s rates, CoreWeave vs Hostinger).
- CoreWeave lists the NVIDIA L40S from $2.25 per GPU-hour on-demand (8-GPU node), $0.98 spot (North America, 8-GPU node). The on-demand rate is unchanged since first read, Sep 7, 2026 (7-day change: none). Against the market: 61% above the median rate card ($1.40 across 38 providers); 34 of the 38 providers we track list a lower on-demand rate, the lowest CloudRift at $0.63 (CloudRift’s rates, CoreWeave vs CloudRift).
- CoreWeave lists the NVIDIA L40 from $1.25 per GPU-hour on-demand (8-GPU node), $0.78 spot (North America, 8-GPU node). The on-demand rate is unchanged since first read, Sep 8, 2026 (7-day change: none). Against the market: 31% above the median rate card ($0.95 across 10 providers); 8 of the 10 providers we track list a lower on-demand rate, the lowest GPU.ai at $0.78 (GPU.ai’s rates, CoreWeave vs GPU.ai).
All listed prices
One table per price type, never mixed. Every rate normalized to one GPU for one hour; a price opens the page it was read from, and 7d marks a rate that moved in the past seven days. The day each figure was first listed is in this page’s data, below.
On-demand
| GPU · what is sold | $ / GPU-hour | Read | Source |
|---|---|---|---|
| NVIDIA L40$10.00/node ÷ 8 | $1.25 | captured | Deploy ↗ |
| NVIDIA L40S$18.00/node ÷ 8 | $2.25 | captured | Deploy ↗ |
| NVIDIA RTX PRO 6000 Blackwell$20.00/node ÷ 8 | $2.50 | captured | Deploy ↗ |
| NVIDIA A100 80GB$21.60/node ÷ 8 | $2.70 | captured | Deploy ↗ |
| NVIDIA H100 SXM$49.24/node ÷ 8 | $6.16 | captured | Deploy ↗ |
| NVIDIA H200$50.44/node ÷ 8 | $6.30 | captured | Deploy ↗ |
| NVIDIA GH200 | $6.50 | captured | Deploy ↗ |
| NVIDIA B200$68.80/node ÷ 8 | $8.60 | captured | Deploy ↗ |
| NVIDIA GB200 NVL72$42.00/node ÷ 4 | $10.50 | captured | Deploy ↗ |
Spot / interruptible
| GPU · what is sold | $ / GPU-hour | Read | Source |
|---|---|---|---|
| NVIDIA L40$6.27/node ÷ 8 · North America · −$0.47 (−37%) vs its on-demand $1.25 | $0.78 | captured | Deploy ↗ |
| NVIDIA L40S$7.88/node ÷ 8 · North America · −$1.27 (−56%) vs its on-demand $2.25 | $0.98 | captured | Deploy ↗ |
| NVIDIA A100 80GB$9.51/node ÷ 8 · Europe · −$1.51 (−56%) vs its on-demand $2.70 | $1.19 | captured | Deploy ↗ |
| NVIDIA RTX PRO 6000 Blackwell$9.56/node ÷ 8 · −$1.30 (−52%) vs its on-demand $2.50 | $1.20 | captured | Deploy ↗ |
| NVIDIA A100 80GB$9.65/node ÷ 8 · North America · −$1.49 (−55%) vs its on-demand $2.70 | $1.21 | captured | Deploy ↗ |
| NVIDIA H100 SXM$19.51/node ÷ 8 · Europe · −$3.72 (−60%) vs its on-demand $6.16 | $2.44 | captured | Deploy ↗ |
| NVIDIA H100 SXM$19.71/node ÷ 8 · North America · −$3.69 (−60%) vs its on-demand $6.16 | $2.46 | captured | Deploy ↗ |
| NVIDIA H200$20.64/node ÷ 8 · Europe · −$3.72 (−59%) vs its on-demand $6.30 | $2.58 | captured | Deploy ↗ |
| NVIDIA H200$20.93/node ÷ 8 · North America · −$3.69 (−59%) vs its on-demand $6.30 | $2.62 | captured | Deploy ↗ |
| NVIDIA B200$34.11/node ÷ 8 · North America · −$4.34 (−50%) vs its on-demand $8.60 | $4.26 | captured | Deploy ↗ |
| NVIDIA B200$34.87/node ÷ 8 · Europe · −$4.24 (−49%) vs its on-demand $8.60 | $4.36 | captured | Deploy ↗ |
| NVIDIA B300$35.84/node ÷ 8 · North America | $4.48 | captured | Deploy ↗ |
| NVIDIA B300$36.70/node ÷ 8 · Europe | $4.59 | captured | Deploy ↗ |
What else CoreWeave bills
The GPU-hour is not the whole bill. What CoreWeave's own pages say about the rest, read by hand and checked on ; each line links the page it was read from. "Not published" means we looked and the pages do not say. Every provider side by side.
| What | What CoreWeave states | Source |
|---|---|---|
| In the rate | Instance price listed with its vCPU, RAM and local NVMe (HGX H100: 128 vCPU, 2,048 GB, 61.44 TB) at $49.24/hr our reading | source checked |
| Data out | Free: no charge for data transfer to the internet | source checked |
| Storage | Distributed File Storage $0.070/GB-month (binary GB); billed on provisioned capacity | source checked |
| While stopped | Not published; nodes have no stopped state on the pricing or billing pages. DFS billed on provisioned capacity | source checked |
| Billing step | Increment and minimum not published; usage counted in instance-hours, attributed every 30 seconds | source checked |
Every rate on this page as data: /providers/coreweave/prices.json — the same fields as prices.json, narrowed to this provider. CC BY 4.0, no key, no rate limit. On your own page: embed this table (/embed/providers/coreweave, an iframe, no script).
Compare CoreWeave with
Every provider that lists at least two of the same GPUs on-demand, most shared GPUs first; each opens the side-by-side page.
- GPU.ai 7 shared GPUs
- Oracle Cloud (OCI) 7 shared GPUs
- RunPod 7 shared GPUs
- Amazon Web Services 6 shared GPUs
- EmpirioLabs 6 shared GPUs
- HexGrid 6 shared GPUs
- Hyperstack 6 shared GPUs
- Modal 6 shared GPUs
- Sesterce 6 shared GPUs
- Vast.ai 6 shared GPUs
- Verda 6 shared GPUs
- Barrack AI 5 shared GPUs
- Beam 5 shared GPUs
- Cerebrium 5 shared GPUs
- Crusoe 5 shared GPUs
- E2E Networks 5 shared GPUs
- Geodd 5 shared GPUs
- Hugging Face 5 shared GPUs
- Koyeb 5 shared GPUs
- Lium 5 shared GPUs
- Massed Compute 5 shared GPUs
- Microsoft Azure 5 shared GPUs
- Nebius 5 shared GPUs
- Seeweb 5 shared GPUs
- Bitdeer AI 4 shared GPUs
- Civo 4 shared GPUs
- Gcore 4 shared GPUs
- Google Cloud 4 shared GPUs
- Hikube 4 shared GPUs
- Hostinger 4 shared GPUs
- Jarvislabs 4 shared GPUs
- Lambda 4 shared GPUs
- Lyceum 4 shared GPUs
- packet.ai 4 shared GPUs
- QuantaCloud 4 shared GPUs
- Cudo Compute 3 shared GPUs
- Dataoorts 3 shared GPUs
- Daytona 3 shared GPUs
- Deep Infra 3 shared GPUs
- Denvr Dataworks 3 shared GPUs
- Elice Cloud 3 shared GPUs
- fal 3 shared GPUs
- FPT AI Factory 3 shared GPUs
- GMI Cloud 3 shared GPUs
- HPC-AI.COM 3 shared GPUs
- Hyperbolic 3 shared GPUs
- Neysa 3 shared GPUs
- OVHcloud 3 shared GPUs
- Paperspace 3 shared GPUs
- Servercore 3 shared GPUs
- STACKIT 3 shared GPUs
- Thunder Compute 3 shared GPUs
- Together AI 3 shared GPUs
- VESSL AI 3 shared GPUs
- Atlas Cloud 2 shared GPUs
- Baseten 2 shared GPUs
- Canopy Wave 2 shared GPUs
- CloudRift 2 shared GPUs
- Fireworks AI 2 shared GPUs
- immers.cloud 2 shared GPUs
- Leafcloud 2 shared GPUs
- Northflank 2 shared GPUs
- Replicate 2 shared GPUs
- RunComfy 2 shared GPUs
- Scaleway 2 shared GPUs
- Selectel 2 shared GPUs
- Taiga Cloud (Northern Data Group) 2 shared GPUs
Data updated: