compare · gpu marketplace
Vast.ai is a peer-to-peer GPU marketplace where independent hosts list machines and renters bid on interruptible or on-demand capacity. GPU.ai takes the other approach: one account that prices 12+ clouds on every launch and passes the raw rate through at zero markup.
at a glance
A fair read of both sides: their genuine strengths, the trade-offs, and what an aggregated market does differently.
Strengths
Trade-offs
The aggregator model
Live per-GPU floor prices across our aggregated fleet. Every row links to the full per-model comparison page.
| GPU | VRAM | GPU.ai / GPU / hr | Stock |
|---|---|---|---|
| RTX A4000 | 16 GB | $0.090/hr | 26 available |
| V100 | 16 GB | $0.090/hr | 2 available |
| RTX 3080 | 10 GB | $0.110/hr | 13 available |
| RTX A4500 | 20 GB | $0.130/hr | 1 available |
| RTX 3090 | 24 GB | $0.150/hr | 34 available |
| RTX 4080 | 16 GB | $0.230/hr | 7 available |
| RTX 2000 Ada | 16 GB | $0.240/hr | 4 available |
| L4 | 24 GB | $0.270/hr | 9 available |
| RTX 5080 | 16 GB | $0.290/hr | 13 available |
| RTX 4090 | 24 GB | $0.340/hr | 88 available |
| RTX A6000 | 48 GB | $0.398/hr | 34 available |
| RTX 5090 | 32 GB | $0.445/hr | 115 available |
| A100 40GB | 40 GB | $0.468/hr | 25 available |
| A40 | 48 GB | $0.490/hr | 8 available |
| RTX 6000 Ada | 48 GB | $0.650/hr | 29 available |
| L40 | 48 GB | $0.690/hr | 10 available |
| RTX PRO 4500 | 32 GB | $0.720/hr | 4 available |
| L40S | 48 GB | $0.790/hr | 137 available |
| A100 80GB | 80 GB | $0.970/hr | 81 available |
| RTX PRO 6000 | 96 GB | $1.69/hr | 89 available |
| H100 PCIe | 80 GB | $1.98/hr | 14 available |
| H100 NVL | 94 GB | $2.59/hr | 15 available |
| H100 SXM | 80 GB | $3.49/hr | 32 available |
| H200 SXM | 141 GB | $4.59/hr | 36 available |
| H200 NVL | 141 GB | $5.47/hr | 2 available |
| B200 | 192 GB | $6.79/hr | 6 available |
| B300 | 288 GB | $7.89/hr | 4 available |
GPU.ai prices are live per-GPU floor rates from our aggregated fleet, normalized per GPU per hour, billed per second, with zero markup on the provider's raw rate. We only show models with launchable stock. Last updated 2026-10-01; this page refreshes hourly.
honest verdict
faq
GPU.ai takes a different approach: instead of running one fleet, it prices the whole GPU market on every launch and routes you to the cheapest provider with live stock, at the provider's raw rate with zero markup. If Vast.ai's strengths (frequently the lowest raw prices on consumer gpus anywhere) are not the thing you're paying for, an aggregated market usually wins on price and availability.
GPU.ai publishes its live per-GPU floor price for every model it carries, currently from $0.090/GPU/hr at the low end. Prices are normalized per GPU per hour and billed per second, and the comparison table on this page is refreshed hourly from live fleet data.
Launches are standard SSH instances with your keys, so most workflows move without changes: pick a GPU, launch, rsync your data, run. There are no commitments to unwind and no minimum spend. You can trial a single instance for a few cents before moving anything.
Honestly: you actively bid-shop and tolerate interruptions for the lowest cost, or if your jobs checkpoint aggressively and restart anywhere. This page's job is a fair comparison, not a hard sell: the live price table above is the argument that matters.