A100 80 GB rental prices per GPU-hour
Ampere, 80 GB. Forms: SXM, PCIe, NVLink bridged. Collected 2026-11-10 from the vendors' own pages. 28 price points, 15 providers.
| Provider | Secure | Community | Market | Spot | Reserved |
|---|---|---|---|---|---|
| RunPod | $1.79 | $1.39 | · | · | · |
| Vast.ai | · | · | from $0.29 | · | · |
| Lambda | $3.19▲ | · | · | · | · |
| Hyperstack | $1.60 | · | · | · | from $1.36 |
| Verda (ex-DataCrunch) | $2.32 | · | · | $1.16 | · |
| CoreWeave | $2.70 | · | · | $1.21 | · |
| Crusoe | $2.30 | · | · | · | · |
| Vultr | $2.80 | · | · | · | · |
| AWS | $3.43 | · | · | · | · |
| Google Cloud | $5.07 | · | · | · | · |
| Microsoft Azure | $4.10 | · | · | · | · |
cooler = cheaper warmer = pricier (scaled within this map, reserved excluded). Purple dashed = "from" price, committed term or announced change. Not every provider has every tier or form.
| Provider | Secure | Community | Market | Spot |
|---|---|---|---|---|
| RunPod | $1.79 | $1.19 | · | · |
| Vast.ai | · | · | from $0.33 | · |
| Hyperstack | $1.35 | · | · | $1.08 |
| Crusoe | $2.00 | · | · | · |
| Vultr | $2.40 | · | · | · |
cooler = cheaper warmer = pricier (scaled within this map, reserved excluded). Purple dashed = "from" price, committed term or announced change. Not every provider has every tier or form.
| Provider | Secure |
|---|---|
| Hyperstack | $1.40 |
cooler = cheaper warmer = pricier (scaled within this map, reserved excluded). Purple dashed = "from" price, committed term or announced change. Not every provider has every tier or form.
All A100 80 GB offers
Rows ordered by price within each tier. The form column is mandatory: compare only matching forms.
| Tier | Provider | Form | $ / GPU-hr | Notes | Source |
|---|---|---|---|---|---|
| Secure | Thunder Compute | form not stated | $1.09 | A100 80GB, per-minute billing | price page |
| Secure | Hyperstack | PCIe | $1.35 | price page | |
| Secure | Hyperstack | NVL / NVLink-bridged | $1.40 | listed as "NVIDIA A100 NVLink" | price page |
| Secure | Jarvis Labs | form not stated | $1.49 | listed as "A100 80GB" | price page |
| Secure | Hyperstack | SXM (HGX-class) | $1.60 | price page | |
| Secure | RunPod | SXM (HGX-class) | $1.79 | price page | |
| Secure | RunPod | PCIe | $1.79 | price page | |
| Secure | Crusoe | PCIe | $2.00 | A100 80GB PCIe | price page |
| Secure | Crusoe | SXM (HGX-class) | $2.30 | A100 80GB SXM | price page |
| Secure | Verda (ex-DataCrunch) | SXM (HGX-class) | $2.316 | price page | |
| Secure | Vultr | PCIe | $2.397 | PCIe A100, single GPU | price page |
| Secure | CoreWeave | SXM (HGX-class) | $2.70 | per GPU-hour from the 8-GPU node price ($21.60) | price page |
| Secure | Vultr | SXM (HGX-class) | $2.80 | HGX A100, 8-GPU node | price page |
| Secure | Lambda | SXM (HGX-class) | $3.19 | 8x HGX A100 80 GB; announced price effective 2026-11-16 | price page |
| Secure | AWS | SXM (HGX-class) | $3.43 | p4de.24xlarge/8, 80GB, us-east-1 | price page |
| Secure | Microsoft Azure | SXM (HGX-class) | $4.10 | ND96amsr A100 v4/8, 80GB, East US | price page |
| Secure | Google Cloud | SXM (HGX-class) | $5.07 | a2-ultragpu-1g, 80GB, us-central1 | price page |
| Community | RunPod | PCIe | $1.19 | price page | |
| Community | RunPod | SXM (HGX-class) | $1.39 | price page | |
| Market | Vast.ai | SXM (HGX-class) | from $0.29 median $0.84 | price page | |
| Market | Vast.ai | PCIe | from $0.33 median $0.73 | price page | |
| Serverless | Modal | form not stated | $2.498 | $0.000694 per second × 3,600 | price page |
| Serverless | Replicate | form not stated | $5.04 | gpu-a100-large, $0.001400 per second | price page |
| Spot | Jarvis Labs | form not stated | $0.89 | price page | |
| Spot | Hyperstack | PCIe | $1.08 | price page | |
| Spot | Verda (ex-DataCrunch) | SXM (HGX-class) | $1.158 | price page | |
| Spot | CoreWeave | SXM (HGX-class) | $1.21 | spot per GPU-hour from the 8-GPU node price ($9.65) | price page |
| Reserved | Hyperstack | SXM (HGX-class) | from $1.36 | reservation "starting from" | price page |
Questions
How much does it cost to rent an A100 80 GB?
On 2026-11-10: secure on-demand SXM-class prices ranged $1.60 to $5.07 per GPU-hour across 10 providers; community and marketplace prices were lower and spot lower still. Form and tier matter, see the table.
What is the difference between SXM and PCIe A100 80 GB?
SXM cards sit on HGX boards with a fast GPU-to-GPU fabric and are usually priced higher; PCIe cards are standard add-in cards. We never rank them against each other without a label.
Is spot cheaper than on-demand?
Per hour, yes, but spot can be interrupted. The estimator adds your own rerun rate so you can compare the whole job.
Sponsored row: none yet. Advertise on this page (waitlist). Sponsored rows would sit above the table and never change the order. Total a job.