Provider record · updated 2026-09-10
Vast.ai (live API) — GPU pricing & deliverability
5 tracked offerings · grade mix A:0 · B:0 · C:1 · D:1 · U:3. Public bundles API exposes price, GPU count, VRAM, geolocation and PCIe details per listing, but nothing about inter-node fabric or NVLink topology. Multi-node contiguity is not a marketplace concept here: each listing is a single host.
Tracked offerings
| GPU | SKU | $/GPU-hr | Terms | Grade | Fabric (as published) | Src |
|---|---|---|---|---|---|---|
| RTX4090 | RTX 4090 8x (marketplace, cheapest of 13) | $0.37 | on-demand | U | not published | ↗ |
| RTX5090 | RTX 5090 8x (marketplace, cheapest of 44) | $0.40 | on-demand | U | not published | ↗ |
| A100-SXM | A100 SXM4 8x (marketplace, cheapest of 2) | $0.91 | on-demand | C | not published | ↗ |
| H100-PCIe | H100 PCIE 8x (marketplace, cheapest of 3) | $2.40 | on-demand | D | not published | ↗ |
| H200 | H200 8x (marketplace, cheapest of 1) | $4.61 | on-demand | U | not published | ↗ |
What Vast.ai (live API) does not publish
- inter_node_fabric
- oversubscription
- max_contiguous_gpus beyond single host
Unpublished fields are recorded as unknown and grade conservatively. Sellers improve their grades by disclosing — that is the point.
Weekly context on every provider
The Delivered Compute Report tracks how these prices and grades move.
Subscribe free