GPU cloud · 9 GPU models tracked
Koyeb
$0.50–$5.50per GPU-hour, the cheapest row to the dearest
Serverless GPU instances that scale to zero and are billed by the second, shown here as an hour. Each price covers the whole instance, with CPU, memory and disk included.
As of , Koyeb lists 9 GPU models we track, from $0.50 per GPU-hour (NVIDIA RTX 4000 SFF Ada, serverless) to $5.50 (NVIDIA B200, serverless).
| Website | www.koyeb.com/ |
|---|---|
| Pricing page | www.koyeb.com/pricing |
| Regions and tiers | No region or tier named in the rates below: one global rate card |
| Models tracked | 9 |
Price by GPU
As of , per GPU-hour, cheapest tier first named:
- Koyeb lists the NVIDIA B200 from $5.50 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 16, 2026 (7-day change: none). Against the market: 19% below the median rate card ($6.79 across 27 providers); 4 of the 27 providers we track list a lower on-demand rate, the lowest Deep Infra at $3.69 (Deep Infra’s rates, Koyeb vs Deep Infra).
- Koyeb lists the NVIDIA H200 from $3.00 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 16, 2026 (7-day change: none). Against the market: 32% below the median rate card ($4.38 across 38 providers); 5 of the 38 providers we track list a lower on-demand rate, the lowest Beam at $2.09 (Beam’s rates, Koyeb vs Beam).
- Koyeb lists the NVIDIA H100 80GB (form factor not stated) from $2.50 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 19, 2026 (7-day change: none). Against the market: 10% below the median rate card ($2.79 across 19 providers); 4 of the 19 providers we track list a lower on-demand rate, the lowest Gcore at $1.74 (Gcore’s rates, Koyeb vs Gcore).
- Koyeb lists the NVIDIA RTX PRO 6000 Blackwell from $2.20 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 16, 2026 (7-day change: none). Against the market: at the median rate card ($2.20 across 27 providers); 13 of the 27 providers we track list a lower on-demand rate, the lowest Beam at $1.09 (Beam’s rates, Koyeb vs Beam).
- Koyeb lists the NVIDIA A100 80GB from $1.60 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 16, 2026 (7-day change: none). Against the market: 12% below the median rate card ($1.81 across 40 providers); 15 of the 40 providers we track list a lower on-demand rate, the lowest Deep Infra at $0.89 (Deep Infra’s rates, Koyeb vs Deep Infra).
- Koyeb lists the NVIDIA L40S from $1.20 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 16, 2026 (7-day change: none). Against the market: 14% below the median rate card ($1.40 across 28 providers); 10 of the 28 providers we track list a lower on-demand rate, the lowest Beam at $0.76 (Beam’s rates, Koyeb vs Beam).
- Koyeb lists the NVIDIA RTX A6000 from $0.75 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 16, 2026 (7-day change: none). Against the market: 25% above the median rate card ($0.60 across 12 providers); 7 of the 12 providers we track list a lower on-demand rate, the lowest Thunder Compute at $0.35 (Thunder Compute’s rates, Koyeb vs Thunder Compute).
- Koyeb lists the NVIDIA L4 from $0.70 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 16, 2026 (7-day change: none). Against the market: 12% below the median rate card ($0.80 across 17 providers); 6 of the 17 providers we track list a lower on-demand rate, the lowest Seeweb at $0.43 (Seeweb’s rates, Koyeb vs Seeweb).
- Koyeb lists the NVIDIA RTX 4000 SFF Ada from $0.50 per GPU-hour on-demand. The on-demand rate is unchanged since first read, Sep 25, 2026. Against the market: 6% above the median rate card ($0.47 across 2 providers); 1 of the 2 providers we track list a lower on-demand rate, the lowest RunPod at $0.44 (RunPod’s rates, Koyeb vs RunPod).
All listed prices
One table per price type, never mixed. Every rate normalized to one GPU for one hour.
On-demand
| GPU | USD / GPU-hour | 7-day change | Type | Region | Min. commitment | Source, captured | Deploy |
|---|---|---|---|---|---|---|---|
| NVIDIA RTX 4000 SFF Ada | $0.50 | — | on-demand | — | — | Price page unchanged since first read, Sep 25, 2026 | Deploy ↗ |
| NVIDIA L4 | $0.70 | no change | on-demand | — | — | Price page unchanged since first read, Sep 16, 2026 | Deploy ↗ |
| NVIDIA RTX A6000 | $0.75 | no change | on-demand | — | — | Price page unchanged since first read, Sep 16, 2026 | Deploy ↗ |
| NVIDIA L40S | $1.20 | no change | on-demand | — | — | Price page unchanged since first read, Sep 16, 2026 | Deploy ↗ |
| NVIDIA A100 80GB | $1.60 | no change | on-demand | — | — | Price page unchanged since first read, Sep 16, 2026 | Deploy ↗ |
| NVIDIA RTX PRO 6000 Blackwell | $2.20 | no change | on-demand | — | — | Price page unchanged since first read, Sep 16, 2026 | Deploy ↗ |
| NVIDIA H100 80GB (form factor not stated) | $2.50 | no change | on-demand | — | — | Price page unchanged since first read, Sep 19, 2026 | Deploy ↗ |
| NVIDIA H200 | $3.00 | no change | on-demand | — | — | Price page unchanged since first read, Sep 16, 2026 | Deploy ↗ |
| NVIDIA B200 | $5.50 | no change | on-demand | — | — | Price page unchanged since first read, Sep 16, 2026 | Deploy ↗ |
Every rate on this page as data: /providers/koyeb/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/koyeb, an iframe, no script).
Compare Koyeb 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.
- EmpirioLabs 8 shared GPUs
- RunPod 8 shared GPUs
- Cerebrium 7 shared GPUs
- E2E Networks 7 shared GPUs
- Hugging Face Inference Endpoints 7 shared GPUs
- Vast.ai 7 shared GPUs
- Amazon Web Services 6 shared GPUs
- Beam 6 shared GPUs
- HexGrid 6 shared GPUs
- Modal 6 shared GPUs
- Seeweb 6 shared GPUs
- Verda 6 shared GPUs
- CoreWeave 5 shared GPUs
- Hyperstack 5 shared GPUs
- Lyceum 5 shared GPUs
- Massed Compute 5 shared GPUs
- Oracle Cloud (OCI) 5 shared GPUs
- Baseten 4 shared GPUs
- Daytona 4 shared GPUs
- Deep Infra 4 shared GPUs
- fal 4 shared GPUs
- Gcore 4 shared GPUs
- Google Cloud 4 shared GPUs
- Jarvislabs 4 shared GPUs
- Nebius 4 shared GPUs
- Northflank 4 shared GPUs
- OVHcloud 4 shared GPUs
- packet.ai 4 shared GPUs
- Sesterce 4 shared GPUs
- Civo 3 shared GPUs
- Crusoe 3 shared GPUs
- GMI Cloud 3 shared GPUs
- immers.cloud 3 shared GPUs
- Lambda 3 shared GPUs
- Leafcloud 3 shared GPUs
- Microsoft Azure 3 shared GPUs
- Neysa 3 shared GPUs
- Paperspace 3 shared GPUs
- Replicate 3 shared GPUs
- Selectel 3 shared GPUs
- Cloud.ru Evolution 2 shared GPUs
- Cudo Compute 2 shared GPUs
- Hyperbolic 2 shared GPUs
- Scaleway 2 shared GPUs
- Thunder Compute 2 shared GPUs
- Together AI 2 shared GPUs
Data updated: