The cheapest RTX 3080 on the table right now is $0.060 per GPU-hour at Salad Cloud, out of 7 offers from 2 providers. The RTX 3080 carries 10 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 $0.060 to $0.136 per GPU-hour — a 2.3× 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 $0.060 at Salad Cloud · spot $0.093 at Vast.ai.
| Provider | Tier | $/GPU-hr | Pricing | Node | Commit | In stock | |
|---|---|---|---|---|---|---|---|
| Salad Cloud | marketplace | $0.060 | on-demand | 1× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.093 | spot | 1, 2× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.094 | spot | 4× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.109 | on-demand | 1× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.121 | on-demand | 2× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.136 | on-demand | 4× GPU | — | yes | rent |
How much does an RTX 3080 cost per hour?
As of the latest scrape, $0.060 per GPU-hour at Salad Cloud is the cheapest RTX 3080 offer across 2 tracked cloud providers. The cheapest on-demand rate is $0.060 per GPU-hour at Salad Cloud, and the most expensive on-demand rate is $0.136 at Vast.ai.
Which cloud is cheapest for the RTX 3080?
Salad Cloud at $0.060 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 RTX 3080 capacity cheaper?
Yes — the cheapest spot RTX 3080 is $0.093 per GPU-hour at Vast.ai, +55% versus the cheapest on-demand rate of $0.060. 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 RTX 3080 have?
10 GB, on the Ampere architecture. Multi-GPU instances multiply that: an 8× RTX 3080 node exposes 80 GB of GPU memory in total.