GPU cloud · 7 GPU models tracked
Crusoe
$1.50–$8.96per GPU-hour, the cheapest row to the dearest
On-demand rates per GPU-hour. Spot rates and some Blackwell and AMD rates are available by quote only and are not listed.
As of , Crusoe lists 7 GPU models we track, from $1.50 per GPU-hour (NVIDIA L40S, on-demand) to $8.96 (NVIDIA B300, on-demand).
| Website | www.crusoe.ai/ |
|---|---|
| Pricing page | www.crusoe.ai/cloud/pricing |
| Regions and tiers | No region or tier named in the rates below: one global rate card |
| Models tracked | 7 |
Price by GPU
As of , per GPU-hour: Crusoe’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 | Crusoe on-demand, per GPU-hour | Against the market (median rate card) |
|---|---|---|
| NVIDIA B300 | $8.96 captured Deploy ↗ | at the median $8.99 · 24 providers 10 of 24 list lower · lowest Deep Infra $5.89 · compare |
| NVIDIA H200 | $4.29 captured Deploy ↗ | 6% below the median $4.54 · 51 providers 20 of 51 list lower · lowest Beam $2.09 from · compare |
| NVIDIA H100 SXM | $3.90 captured Deploy ↗ | 11% above the median $3.52 · 38 providers 22 of 38 list lower · lowest Cudo Compute $1.79 GPU only · compare |
| AMD Instinct MI300X | $3.45 captured Deploy ↗ | 23% above the median $2.81 · 8 providers 5 of 8 list lower · lowest TensorWave $1.71 from |
| NVIDIA A100 80GB | $2.00 captured Deploy ↗ | at the median $2.00 · 58 providers 29 of 58 list lower · lowest CloudRift $1.05 · compare |
| NVIDIA L40S | $1.50 captured Deploy ↗ | 8% above the median $1.40 · 38 providers 19 of 38 list lower · lowest CloudRift $0.63 · compare |
The same, in words
- Crusoe lists the NVIDIA B300 from $8.96 per GPU-hour on-demand. The on-demand rate is unchanged since Oct 22, 2026. Against the market: at the median rate card ($8.99 across 24 providers); 10 of the 24 providers we track list a lower on-demand rate, the lowest Deep Infra at $5.89 (Deep Infra’s rates, Crusoe vs Deep Infra).
- Crusoe lists the NVIDIA H200 from $4.29 per GPU-hour on-demand. The on-demand rate is unchanged since Oct 21, 2026 (7-day change: −20.6%). Against the market: 6% below the median rate card ($4.54 across 51 providers); 20 of the 51 providers we track list a lower on-demand rate, the lowest Beam at $2.09 (Beam’s rates, Crusoe vs Beam).
- Crusoe lists the NVIDIA H100 SXM from $3.90 per GPU-hour on-demand. The on-demand rate is unchanged since Oct 21, 2026 (7-day change: −13.3%). Against the market: 11% above the median rate card ($3.52 across 38 providers); 22 of the 38 providers we track list a lower on-demand rate, the lowest Cudo Compute at $1.79 (Cudo Compute’s rates, Crusoe vs Cudo Compute).
- Crusoe lists the AMD Instinct MI300X from $3.45 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 18, 2026 (7-day change: none). Against the market: 23% above the median rate card ($2.81 across 8 providers); 5 of the 8 providers we track list a lower on-demand rate, the lowest TensorWave at $1.71 (TensorWave’s rates).
- Crusoe lists the NVIDIA A100 80GB from $2.00 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 13, 2026 (7-day change: none). Against the market: at the median rate card ($2.00 across 58 providers); 29 of the 58 providers we track list a lower on-demand rate, the lowest CloudRift at $1.05 (CloudRift’s rates, Crusoe vs CloudRift).
- Crusoe lists the NVIDIA L40S from $1.50 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 13, 2026 (7-day change: none). Against the market: 8% above the median rate card ($1.40 across 38 providers); 19 of the 38 providers we track list a lower on-demand rate, the lowest CloudRift at $0.63 (CloudRift’s rates, Crusoe vs CloudRift).
Dated pieces on Crusoe: — Crusoe H100 back to $3.90 an hour, H200 $4.29, October 21, 2026 · — Crusoe H100 went $3.90 to $4.50 an hour, H200 $5.40, October 15, 2026. Each is frozen as of its date.
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 L40S | $1.50 | captured | Deploy ↗ |
| NVIDIA A100 80GB | $2.00 | captured | Deploy ↗ |
| AMD Instinct MI300X | $3.45 | captured | Deploy ↗ |
| NVIDIA H100 SXM7d −13.3% | $3.90 | captured | Deploy ↗ |
| NVIDIA H2007d −20.6% | $4.29 | captured | Deploy ↗ |
| NVIDIA B200 | $8.50 | unconfirmedlast confirmed | Deploy ↗ |
| NVIDIA B300 | $8.96 | captured | Deploy ↗ |
What else Crusoe bills
The GPU-hour is not the whole bill. What Crusoe'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 Crusoe states | Source |
|---|---|---|
| In the rate | Local NVMe (ephemeral) included; instance types bundle vCPU and RAM; persistent and OS disks billed separately our reading | source checked |
| Data out | free: no charge for ingress or egress, in VPC or to/from internet | source checked |
| Storage | Persistent disk $0.08/GiB-month (shared disks $0.07) | source checked |
| While stopped | Stopped on-demand VM: no compute charge; 128 GB OS disk billed at $0.08/GiB-month; reserved VMs bill regardless | source checked |
| Billing step | Per second while running (pricing FAQ says per-hour or per-second); no minimum commitment | source checked |
Every rate on this page as data: /providers/crusoe/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/crusoe, an iframe, no script).
Compare Crusoe 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.
- Oracle Cloud (OCI) 7 shared GPUs
- RunPod 7 shared GPUs
- Amazon Web Services 6 shared GPUs
- GPU.ai 6 shared GPUs
- Modal 6 shared GPUs
- Verda 6 shared GPUs
- Barrack AI 5 shared GPUs
- Civo 5 shared GPUs
- CoreWeave 5 shared GPUs
- EmpirioLabs 5 shared GPUs
- Gcore 5 shared GPUs
- HexGrid 5 shared GPUs
- Hyperstack 5 shared GPUs
- Lium 5 shared GPUs
- Lyceum 5 shared GPUs
- Nebius 5 shared GPUs
- Paperspace 5 shared GPUs
- Seeweb 5 shared GPUs
- Vast.ai 5 shared GPUs
- Beam 4 shared GPUs
- Cerebrium 4 shared GPUs
- Deep Infra 4 shared GPUs
- E2E Networks 4 shared GPUs
- Geodd 4 shared GPUs
- Google Cloud 4 shared GPUs
- Hugging Face 4 shared GPUs
- Koyeb 4 shared GPUs
- Microsoft Azure 4 shared GPUs
- Sesterce 4 shared GPUs
- Together AI 4 shared GPUs
- Bitdeer AI 3 shared GPUs
- Cudo Compute 3 shared GPUs
- Daytona 3 shared GPUs
- Elice Cloud 3 shared GPUs
- fal 3 shared GPUs
- Fireworks AI 3 shared GPUs
- FPT AI Factory 3 shared GPUs
- Hikube 3 shared GPUs
- Hostinger 3 shared GPUs
- HPC-AI.COM 3 shared GPUs
- Hyperbolic 3 shared GPUs
- Jarvislabs 3 shared GPUs
- Lambda 3 shared GPUs
- Massed Compute 3 shared GPUs
- Neysa 3 shared GPUs
- Northflank 3 shared GPUs
- OVHcloud 3 shared GPUs
- packet.ai 3 shared GPUs
- Scaleway 3 shared GPUs
- STACKIT 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
- Dataoorts 2 shared GPUs
- Denvr Dataworks 2 shared GPUs
- GMI Cloud 2 shared GPUs
- immers.cloud 2 shared GPUs
- QuantaCloud 2 shared GPUs
- Replicate 2 shared GPUs
- RunComfy 2 shared GPUs
- Servercore 2 shared GPUs
- Taiga Cloud (Northern Data Group) 2 shared GPUs
- Thunder Compute 2 shared GPUs
Data updated: