The cheapest A10 on the table right now is $0.201 per GPU-hour at Vast.ai, out of 8 offers from 5 providers. The A10 carries 24 GB of VRAM on NVIDIA's Ampere generation. That is peer/marketplace capacity — rented from other users' machines, and not directly comparable to a dedicated instance.
On-demand pricing for the same card spans $0.201 to $1.29 per GPU-hour — a 6.4× spread between the cheapest and the most expensive provider for identical silicon. That gap is the entire reason this table exists.
Cheapest by pricing model: on-demand $0.201 at Vast.ai · spot $0.227 at Vast.ai.
| Provider | Tier | $/GPU-hr | Pricing | Node | Commit | In stock | |
|---|---|---|---|---|---|---|---|
| Vast.ai | marketplace | $0.201 | on-demand | 2× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.227 | spot | 2× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.240 | spot | 1× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.243 | on-demand | 1× GPU | — | yes | rent |
| OVHcloud | dedicated | $1.05 | on-demand | 1× GPU | — | — | rent |
| Modal | dedicated | $1.10 | on-demand | 1× GPU | — | yes | rent |
| Lambda Labs | dedicated | $1.29 | on-demand | 1× GPU | — | — | rent |
| Prime Intellect | dedicated | $1.29 | on-demand | 1× GPU | — | yes | rent |
How much does an A10 cost per hour?
As of the latest scrape, $0.201 per GPU-hour at Vast.ai is the cheapest A10 offer across 5 tracked cloud providers. The cheapest on-demand rate is $0.201 per GPU-hour at Vast.ai, and the most expensive on-demand rate is $1.29 at Prime Intellect.
Which cloud is cheapest for the A10?
Vast.ai at $0.201 per GPU-hour. Prices move constantly, so this page is regenerated from the live scrape every 15 minutes. Note that marketplace tiers are peer hardware and are not directly comparable to dedicated capacity.
Is spot A10 capacity cheaper?
Yes — the cheapest spot A10 is $0.227 per GPU-hour at Vast.ai, +13% versus the cheapest on-demand rate of $0.201. Spot instances can be reclaimed by the provider at any time, so they suit checkpointed training and batch inference rather than long-lived services.
How many GB of VRAM does the A10 have?
24 GB, on the Ampere architecture. Multi-GPU instances multiply that: an 8× A10 node exposes 192 GB of GPU memory in total.