RTX 4090 · 24 GB · Ada Lovelace
RTX 4090 rental prices across providers
As of 12 November 2026, the lowest listed on-demand price for a RTX 4090 among 8 providers is $0.33 per GPU-hour at Novita AI. The median of each provider's lowest price is $0.43 and the highest is $0.89.
- Lowest listed
- $0.33
- Median
- $0.43
- Highest
- $0.89
- Providers
- 8
Vast.ai, per GPU-hour
of each provider’s lowest price
Runcrate
22 configurations
By provider
RTX 4090 price by provider
- Vast.ai$0.13/GPU-hr
- Novita AI$0.33/GPU-hr
- RunPod$0.34/GPU-hr
- TensorDock$0.35/GPU-hr
- packet.ai$0.39/GPU-hr
- gpu.ai$0.40/GPU-hr
- Nova Cloud$0.42/GPU-hr
- RunC.AI$0.42/GPU-hr
- Beam$0.44/GPU-hr
- SaladCloud$0.44/GPU-hr
- CloudRift$0.45/GPU-hr
- SynpixCloud$0.47/GPU-hr
- Spheron$0.48/GPU-hr
- Runcrateours$0.58/GPU-hr
Every listing
All RTX 4090 configurations we track
| Configuration | Billing | Region | Source | ||||
|---|---|---|---|---|---|---|---|
| Vast.aimarketplace | 1× · Marketplace lowest listed (RTX 4090) | not stated | $0.13 | $0.13 | not stated | 2026-11-12 | page |
| Novita AI | 1× · GPU instance, on-demand (sample configuration) | not stated | $0.33 | $0.33 | not stated | 2026-11-12 | page |
| RunPodmarketplace | 1× · Community Cloud RTX 4090 | not stated | $0.34 | $0.34 | not stated | 2026-11-12 | page |
| TensorDockmarketplace | 1× · Marketplace from-price (RTX 4090) | not stated | $0.35 | $0.35 | not stated | 2026-11-12 | page |
| packet.ai | 1× · Dedicated (whole card) | per hour | $0.39 | $0.39 | not stated | 2026-11-12 | page |
| gpu.aimarketplace | 1× · Community Cloud live floor (RTX 4090) | per second | $0.40 | $0.40 | not stated | 2026-11-12 | page |
| gpu.aimarketplace | 8× · 8x RTX 4090 node live floor | per second | $0.41 | $3.29 | not stated | 2026-11-12 | page |
| Nova Cloud | 1× · 1x RTX 4090, 64 GB system RAM | per second | $0.42 | $0.42 | not stated | 2026-11-12 | page |
| Nova Cloud | 2× · 2x RTX 4090, 128 GB system RAM | per second | $0.42 | $0.84 | not stated | 2026-11-12 | page |
| Nova Cloud | 4× · 4x RTX 4090, 256 GB system RAM | per second | $0.42 | $1.68 | not stated | 2026-11-12 | page |
| Nova Cloud | 8× · 8x RTX 4090, 512 GB system RAM | per second | $0.42 | $3.36 | not stated | 2026-11-12 | page |
| RunC.AI | 1× · 1x RTX 4090 | not stated | $0.42 | $0.42 | not stated | 2026-11-12 | page |
| RunC.AI | 2× · 2x RTX 4090 | not stated | $0.42 | $0.84 | not stated | 2026-11-12 | page |
| RunC.AI | 3× · 3x RTX 4090 | not stated | $0.42 | $1.26 | not stated | 2026-11-12 | page |
| RunC.AI | 4× · 4x RTX 4090 | not stated | $0.42 | $1.68 | not stated | 2026-11-12 | page |
| RunC.AI | 5× · 5x RTX 4090 | not stated | $0.42 | $2.10 | not stated | 2026-11-12 | page |
| RunC.AI | 6× · 6x RTX 4090 | not stated | $0.42 | $2.52 | not stated | 2026-11-12 | page |
| RunC.AI | 7× · 7x RTX 4090 | not stated | $0.42 | $2.94 | not stated | 2026-11-12 | page |
| RunC.AI | 8× · 8x RTX 4090 | not stated | $0.42 | $3.36 | not stated | 2026-11-12 | page |
| Beammarketplace | 1× PCIe · On-demand machine (from price) | not stated | $0.44 | $0.44 | not stated | 2026-11-12 | page |
| SaladCloudmarketplace | 1× · High priority tier | per second | $0.44 | $0.44 | not stated | 2026-11-12 | page |
| CloudRift | 1× · On-demand (max 6 GPUs per instance) | per second | $0.45 | $0.45 | not stated | 2026-11-12 | page |
| SynpixCloud | 8× · 8x RTX 4090 | not stated | $0.47 | $3.72 | not stated | 2026-11-12 | page |
| SynpixCloud | 6× · 6x RTX 4090 | not stated | $0.47 | $2.81 | not stated | 2026-11-12 | page |
| SynpixCloud | 4× · 4x RTX 4090 | not stated | $0.47 | $1.88 | not stated | 2026-11-12 | page |
| Spheronmarketplace | 1× · Marketplace lowest-priced live on-demand (RTX 4090) | per minute | $0.48 | $0.48 | not stated | 2026-11-12 | page |
| SynpixCloud | 1× · 1x RTX 4090 | not stated | $0.48 | $0.48 | not stated | 2026-11-12 | page |
| SynpixCloud | 2× · 2x RTX 4090 | not stated | $0.48 | $0.96 | not stated | 2026-11-12 | page |
| Vast.aimarketplace | 1× · Marketplace median (RTX 4090) | not stated | $0.48 | $0.48 | not stated | 2026-11-12 | page |
| Runcrateours, live | 1× | per second | $0.58 | $0.58 | 4 regions | 2026-11-12 | page |
| RunPod | 1× · Secure Cloud RTX 4090 | per second | $0.89 | $0.89 | not stated | 2026-11-12 | page |
| gpu.aimarketplace | 1× · Non-community row live floor (RTX 4090) | per second | $0.98 | $0.98 | not stated | 2026-11-12 | page |
Questions
RTX 4090 rental: common questions
What is the lowest RTX 4090 price per hour?
As of 12 November 2026, the lowest listed on-demand price for a RTX 4090 among 8 providers is $0.33 per GPU-hour at Novita AI. The median of each provider's lowest price is $0.43 and the highest is $0.89.
How much does an 8-GPU RTX 4090 server cost per hour?
The lowest 8-GPU on-demand configuration in our data is gpu.ai at $0.41 per GPU, $3.29 per hour for the server (read 2026-11-12).
Which providers bill RTX 4090 time by the second?
gpu.ai, Nova Cloud, SaladCloud, CloudRift, and Runcrate state per-second billing for their lowest-priced RTX 4090 listing.
Is the price per GPU or per server?
Per GPU per hour. The total column multiplies by the number of GPUs in that configuration. Multi-GPU configurations often cost a little more per GPU.
How current is this data?
Prices are read from each provider's public pricing page; the table shows the date each was read, and the latest read is 12 November 2026. Runcrate's own price comes live from runcrate.ai.
Billing as stated on each provider's page, per second: Nova Cloud, CloudRift, Runcrate, and RunPod; per hour: packet.ai. Providers that do not state it are not listed here.
Disclosure
Provider Index is published by the Runcrate team, who also run Runcrate, an AI cloud. Runcrate is listed in the table for RTX 4090, marked "ours, live". It is sorted by price like every other row, with no pinned position. Other providers' prices are read from their own public pricing pages. We have no paid placement.