A100 SXM pricing on Koyeb
Every Koyeb A100 SXM offering we track, normalized to a per-GPU hourly rate and ranked against 26 providers with current A100 SXM pricing.
The cheapest A100 SXM we currently track is $0.580/hr on Denvr Dataworks.
Koyeb A100 SXM rates
Hourly prices per GPU. Multi-GPU rows show the per-GPU rate, not the instance total.
| Price / GPU-hr | Type | GPUs |
|---|---|---|
| $1.60 | On-demand | 1 |
| $1.60 | On-demand | 2 |
| $1.60 | On-demand | 4 |
| $1.60 | On-demand | 8 |
A100 SXM specifications
- VRAM
- 80GB
- Architecture
- Ampere
- CUDA Cores
- 6,912
- Tensor Cores
- 432
- Memory Bandwidth
- 2039 GB/s
- TDP
- 400W
A100 SXM on other providers
- Denvr Dataworks$0.580/hr
- Verda$0.626/hr
- Civo$0.790/hr
- Deep Infra$0.890/hr
- Thunder Compute$1.09/hr
- Seeweb$1.14/hr
- CoreWeave$1.21/hr
- Massed Compute$1.24/hr
- IO.NET$1.29/hr
- Beam$1.30/hr
- Runpod$1.39/hr
- Hyperstack$1.40/hr
- Crusoe$1.45/hr
- Oblivus$1.47/hr
- Sesterce$1.49/hr
- Jarvis Labs$1.49/hr
- AceCloud$1.81/hr
- Lambda Labs$1.99/hr
- Modal$2.10/hr
- Together AI$2.59/hr
- Vultr$2.60/hr
- Amazon AWS$2.74/hr
- Paperspace$3.11/hr
- Microsoft Azure$3.67/hr
- Oracle Cloud$4.00/hr
About Koyeb
Koyeb is a serverless platform offering on-demand NVIDIA GPUs from RTX-class up through B200 with per-second billing, scale-to-zero and global edge deployment. Koyeb is joining Mistral AI to build the future of AI infrastructure.
Frequently Asked Questions
How much does the A100 SXM cost on Koyeb?
Koyeb lists the A100 SXM from $1.60 per GPU-hour across 4 tracked configurations. Prices are collected daily — see the table above for the current rates.
Is Koyeb the cheapest place to rent the A100 SXM?
Koyeb ranks #17 of 26 providers we track with current A100 SXM pricing. The cheapest right now is Denvr Dataworks. Price is only one factor — region availability, network, and storage costs differ between providers.
Which A100 SXM configurations does Koyeb offer?
We currently track 4 A100 SXM offerings from Koyeb, covering the GPU counts, pricing types, and regions listed in the table above. Every price is normalized to a per-GPU hourly rate so it can be compared across providers.