The cheapest L40 on the table right now is $0.320 per GPU-hour at Vast.ai (spot), out of 15 offers from 8 providers. The L40 carries 48 GB of VRAM on NVIDIA's Ada 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 $0.330 to $1.25 per GPU-hour — a 3.8× 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 $0.330 at Lium.io · spot $0.320 at Vast.ai.
| Provider | Tier | $/GPU-hr | Pricing | Node | Commit | In stock | |
|---|---|---|---|---|---|---|---|
| Vast.ai | marketplace | $0.320 | spot | 1, 2× GPU | — | yes | rent |
| Lium.io | marketplace | $0.330 | on-demand | 1, 2× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.335 | on-demand | 2× GPU | — | yes | rent |
| Vast.ai | marketplace | $0.337 | on-demand | 1× GPU | — | yes | rent |
| Runpod Community | marketplace | $0.690 | on-demand | 1× GPU | — | yes | rent |
| Runpod Community | marketplace | $0.690 | spot | 1× GPU | — | yes | rent |
| Thunder Compute | dedicated | $0.790 | on-demand | 1× GPU | — | yes | rent |
| Massed Compute | dedicated | $0.860 | on-demand | 1–4× GPU | — | — | rent |
| Prime Intellect | dedicated | $0.860 | on-demand | 1× GPU | — | yes | rent |
| Hyperstack | dedicated | $1.00 | on-demand | 1× GPU | — | — | rent |
| CoreWeave | dedicated | $1.25 | on-demand | 8× GPU | — | — | rent |
How much does an L40 cost per hour?
As of the latest scrape, $0.320 per GPU-hour at Vast.ai is the cheapest L40 offer across 8 tracked cloud providers. The cheapest on-demand rate is $0.330 per GPU-hour at Lium.io, and the most expensive on-demand rate is $1.25 at CoreWeave.
Which cloud is cheapest for the L40?
Vast.ai at $0.320 per GPU-hour on spot capacity. 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 L40 capacity cheaper?
Yes — the cheapest spot L40 is $0.320 per GPU-hour at Vast.ai, -3% versus the cheapest on-demand rate of $0.330. 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 L40 have?
48 GB, on the Ada architecture. Multi-GPU instances multiply that: an 8× L40 node exposes 384 GB of GPU memory in total.