Replicate vs Vast.ai
As of , Replicate and Vast.ai share 2 GPUs today, but on none do both list an on-demand rate: at least one side is community capacity or a marketplace floor, compared below and not counted. On the priciest shared card, NVIDIA A100 80GB: Replicate $5.04 vs Vast.ai $1.00 (Multi-GPU machines · floor) per GPU-hour — not the same market, not counted.
On-demand, per GPU-hour
| GPU |  Replicate |  Vast.ai | Difference | Cheaper |
|---|---|---|---|---|
| NVIDIA A100 80GB median $1.99 · 63 providers · lowest Yotta Labs $0.92 | $5.04 captured Deploy ↗ | $1.00 Multi-GPU machines · floor captured Deploy ↗ | — | not the same market |
| NVIDIA L40S median $1.42 · 40 providers · lowest CloudRift $0.63 | $3.51 captured Deploy ↗ | $0.73 Multi-GPU machines · floor captured Deploy ↗ | — | not the same market |
How this is compared
Cloud GPU rental prices of Replicate and Vast.ai, compared per GPU-hour on the GPUs both list. Where a provider has several tiers or regions, its cheapest on-demand rate is used and labelled. Community capacity (RunPod Community Cloud) and marketplace floors are shown only where a provider has nothing else, and never decide which is cheaper.
GPU by GPU
- NVIDIA A100 80GB: Replicate $5.04 vs Vast.ai $1.00 (Multi-GPU machines · floor) per GPU-hour — not the same market, not counted.
- NVIDIA L40S: Replicate $3.51 vs Vast.ai $0.73 (Multi-GPU machines · floor) per GPU-hour — not the same market, not counted.
Not compared: GPUs only one of the two lists, reserved/committed tiers, and anything either provider prices only on request. See each provider's page for its full list.
Data updated: