Loading page
RTX 4090 pricing on Vast.ai
Every Vast.ai RTX 4090 offering we track, normalized to a per-GPU hourly rate and ranked against 22 providers with current RTX 4090 pricing.
From / GPU-hr
$0.383
Price rank
#8 of 22
Configurations
4
Last updated
September 19, 2026
The cheapest RTX 4090 we currently track is $0.140/hr on AtmosCompute.
Vast.ai RTX 4090 rates
Hourly prices per GPU. Multi-GPU rows show the per-GPU rate, not the instance total.
| Price / GPU-hr | Type | GPUs |
|---|---|---|
| $0.383 | On-demand | 1 |
| $0.454 | On-demand | 2 |
| $0.521 | On-demand | 8 |
| $0.735 | On-demand | 4 |
RTX 4090 specifications
- VRAM
- 24GB
- Architecture
- Ada Lovelace
- CUDA Cores
- 16,384
- Tensor Cores
- 512
- Memory Bandwidth
- 1008 GB/s
- TDP
- 450W
RTX 4090 on other providers
- AtmosCompute$0.140/hr
- Salad Cloud$0.160/hr
- QuickPod$0.300/hr
- Lium.io$0.330/hr
- Novita AI$0.335/hr
- Runpod$0.340/hr
- 1Legion$0.380/hr
- Packet AI$0.390/hr
- GPU Outlet$0.450/hr
- Hexgrid Cloud$0.450/hr
- gpu.ai$0.481/hr
- SwissGPU$0.484/hr
- Shadeform$0.490/hr
- IO.NET$0.514/hr
- Omega Gradient$0.520/hr
- Sesterce$0.539/hr
- Daytona$0.570/hr
- Oblivus$0.640/hr
- Runcrate$0.660/hr
- Beam$0.690/hr
- Spheron$0.720/hr
About Vast.ai
Vast.ai is a GPU marketplace with live, supply-and-demand pricing across community hosts and Secure Cloud datacenter partners. Instances are provisioned through the console, CLI, Python SDK, or REST API and billed per second.
Frequently Asked Questions
How much does the RTX 4090 cost on Vast.ai?
Vast.ai lists the RTX 4090 from $0.383 per GPU-hour across 4 tracked configurations. Prices are collected daily — see the table above for the current rates.
Is Vast.ai the cheapest place to rent the RTX 4090?
Vast.ai ranks #8 of 22 providers we track with current RTX 4090 pricing. The cheapest right now is AtmosCompute. Price is only one factor — region availability, network, and storage costs differ between providers.
Which RTX 4090 configurations does Vast.ai offer?
We currently track 4 RTX 4090 offerings from Vast.ai, covering the GPU counts, pricing types, and regions listed in the table above. Every price is normalized to a per-GPU hourly rate so it can be compared across providers.