H100 rental price
median price per GPU-hour$2.55 / GPU-hr
Live chart · updates with every published snapshot
Cheapest GPU rental price
The lowest median hourly rental price across every GPU configuration listed on the Akash open compute marketplace, and which configuration holds it.
lowest median price per GPU-hour$0.05 / GPU-hrfor the NVIDIA GTX1050 2 GB PCIe
Behind the number
The full field
- NVIDIA GTX1050 2 GB PCIe$0.05
- NVIDIA GTX1070TI 8 GB PCIe$0.08
- NVIDIA RTX 3060 12 GB PCIe$0.14
- NVIDIA RTX 3090 24 GB PCIe$0.15
- NVIDIA RTX 4080 16 GB PCIe$0.15
- NVIDIA RTX 6000 24 GB PCIe$0.17
- NVIDIA V100 32 GB PCIe$0.21
- NVIDIA RTX 8000 48 GB PCIe$0.21
- NVIDIA P2000 5 GB PCIe$0.30
- NVIDIA RTX 4090 24 GB PCIe$0.34
- NVIDIA RTX 4000ADA 20 GB PCIe$0.41
- NVIDIA RTX 5090 32 GB PCIe$0.64
The board in rank order. The headline number is the top-ranked row, highlighted. Showing the top 12 of 17 ranked entries.
| Rank | Entry | lowest median price per GPU-hour |
|---|---|---|
| 1 | NVIDIA GTX1050 2 GB PCIe | $0.05 / GPU-hr |
| 2 | NVIDIA GTX1070TI 8 GB PCIe | $0.08 / GPU-hr |
| 3 | NVIDIA RTX 3060 12 GB PCIe | $0.14 / GPU-hr |
| 4 | NVIDIA RTX 3090 24 GB PCIe | $0.15 / GPU-hr |
| 5 | NVIDIA RTX 4080 16 GB PCIe | $0.15 / GPU-hr |
| 6 | NVIDIA RTX 6000 24 GB PCIe | $0.17 / GPU-hr |
| 7 | NVIDIA V100 32 GB PCIe | $0.21 / GPU-hr |
| 8 | NVIDIA RTX 8000 48 GB PCIe | $0.21 / GPU-hr |
| 9 | NVIDIA P2000 5 GB PCIe | $0.30 / GPU-hr |
| 10 | NVIDIA RTX 4090 24 GB PCIe | $0.34 / GPU-hr |
| 11 | NVIDIA RTX 4000ADA 20 GB PCIe | $0.41 / GPU-hr |
| 12 | NVIDIA RTX 5090 32 GB PCIe | $0.64 / GPU-hr |
$0.05 / GPU-hr on every publication so far. A line appears here as soon as the number moves.
About this metric
Every market has a floor, and on the open GPU market the floor is where experimentation lives. This chart tracks the single cheapest median hourly rental price across all GPU configurations listed on Akash, an open marketplace whose full order book is public, along with the configuration that currently holds the bottom slot.
Each configuration on the underlying board carries a median price per GPU-hour computed from its active positive-price listings; this chart reads the lowest of those medians at every published snapshot. Zero-price and unavailable listings are excluded, so the floor is a real, rentable offer rather than a promotional artifact.
The price floor is where students, hobbyists, and bootstrapped teams actually buy compute. A falling floor means yesterday's data-center hardware is flooding the open market and the cost of entry to GPU work is dropping. Read it against the H100 line: the gap between the floor and the flagship is the market's price on being at the frontier.
Related charts