NVIDIA GeForce RTX 5090 rental pricing
Blackwell architecture · 32 GB GDDR7 memory · 1.8 TB/s bandwidth · 575 W board power · Released 2025
Price history by provider
Every provider's cheapest daily on-demand rate for the NVIDIA GeForce RTX 5090, last 30 days
One line per provider, reconstructed from recorded price changes for listings currently live. Click a name in the key to hide or show it.
At a glance
prices checked just now- Cheapest in stock
- $0.48/hr
- Community · 4× GeForce RTX 5090 32GB · Vast.ai
- Median price
- $1.02/hr
- On-demand · 4 providers
- 90-day trend
- Up 16.7%
- Weekly median on-demand, 4 weeks
- Coverage
- 7 providers
- 10 price points
- $ per FP8 PFLOP-hour
- $0.86
- Rank 3 of 30 priced GPUs · lower is better
Every provider we track
Cheapest listing per provider and pricing type. Sold-out rows stay so you can see who normally carries it.
| Provider | Configuration | Price/GPU-hr | Checked | |
|---|---|---|---|---|
| 4× GeForce RTX 5090 32GB Community | $0.48 | just now | View → | |
| 1× GeForce RTX 5090 32GB Community | $0.69 | just now | View → | |
 Beam | 1× GeForce RTX 5090 32GB | $0.72 | just now | View → |
| 1× GeForce RTX 5090 32GB Spot | $0.74 | just now | View → | |
 Spheron | 1× GeForce RTX 5090 32GB | $0.84 | just now | View → |
 Lium | 1× GeForce RTX 5090 32GB Community | $0.84 | just now | View → |
| 1× GeForce RTX 5090 32GB | $1.19 | just now | View → | |
| 1× GeForce RTX 5090 32GB | $1.29 | just now | View → | |
 Shadeform Excess Supply capacity | 8× GeForce RTX 5090 32GB | $0.65 | just now | View → |
Per GPU-hour. Spot, reserved and community pricing differ from on-demand; click through for the provider's current figure.
What can the NVIDIA GeForce RTX 5090 run?
Memory for the weights plus one sequence of context, how many GPUs hold it, and what that replica costs at the cheapest in-stock on-demand rate.
| Model | Params | Memory | GPUs | $/hr |
|---|---|---|---|---|
| gpt-oss-20b OpenAI | 20.9B· 3.6B active | 20 GiB | 1 | $0.72 |
| Gemma 3 27B Google | 27.4B | 26 GiB | 1 | $0.72 |
| Qwen3-32B Alibaba | 32.8B | 32 GiB | 2 | $1.44 |
| Llama 3.3 70B Instruct Meta | 70.6B | 67 GiB | 4 | $2.88 |
| gpt-oss-120b OpenAI | 117B· 5.1B active | 109 GiB | 4 | $2.88 |
| Qwen3-235B-A22B Alibaba | 235B· 22.0B active | 220 GiB | 8 | $5.76 |
| DeepSeek-R1 DeepSeek | 684B· 37.0B active | 638 GiB | Over 16 | — |
| Kimi K3 Moonshot AI | 2.78T· 104B active | 2.53 TiB | Over 16 | — |
Sized for one sequence at 4K context with 16-bit KV cache and 10% of memory held back. Serving many users at once needs more; the LLM GPU calculator sizes for concurrency and throughput.
NVIDIA GeForce RTX 5090 pricing questions
How much does it cost to rent a NVIDIA GeForce RTX 5090 per hour?
The cheapest NVIDIA GeForce RTX 5090 in stock right now is $0.48 per GPU-hour community from Vast.ai for the 4× GeForce RTX 5090 32GB configuration. The median on-demand rate across 4 providers is $1.02 per GPU-hour.
How much does a NVIDIA GeForce RTX 5090 cost per month?
Running one NVIDIA GeForce RTX 5090 around the clock for a month (730 hours) costs about $351 at the cheapest in-stock rate of $0.48 per hour. At the median on-demand rate it is about $741 a month. Reserved terms usually come in below that; ask the provider for a quote.
Which provider has the cheapest NVIDIA GeForce RTX 5090?
Vast.ai has the cheapest NVIDIA GeForce RTX 5090 in stock right now at $0.48 per GPU-hour community for the 4× GeForce RTX 5090 32GB configuration. That is a community rate; see the provider table for the cheapest on-demand listing. Rates change often, so check the provider table for today's figure before you commit.
Are NVIDIA GeForce RTX 5090 rental prices going up or down?
The weekly median on-demand rate for the NVIDIA GeForce RTX 5090 is up 16.7% over the last 4 weeks, from $0.72 to $0.84 per GPU-hour.
What is the difference between on-demand, spot and reserved NVIDIA GeForce RTX 5090 pricing?
On-demand is pay-as-you-go with no commitment. Spot is spare capacity sold at a discount that the provider can reclaim at short notice. Community is capacity from individual hosts on a marketplace, cheaper and less uniform. This week's median rates for the NVIDIA GeForce RTX 5090 are on-demand $0.84 (5 providers), spot $0.74 (1 provider), community $0.69 (3 providers), per GPU-hour.
How many cloud providers rent the NVIDIA GeForce RTX 5090?
We track 7 providers with 10 live prices for the NVIDIA GeForce RTX 5090. Sold-out listings stay in the table so you can see who normally carries it, but never set the cheapest figure.
What LLMs can you run on one NVIDIA GeForce RTX 5090?
With 8-bit weights and a single sequence of context, one NVIDIA GeForce RTX 5090 holds gpt-oss-20b (20.9B), Gemma 3 27B (27.4B). See the model fit table for the full ladder.
How does the NVIDIA GeForce RTX 5090 compare on price per PFLOP-hour?
At its cheapest in-stock on-demand rate the NVIDIA GeForce RTX 5090 delivers dense FP8 compute at $0.86 per PFLOP-hour, which ranks 3 of 30 priced GPUs we cover on that basis. Lower is better: it is the rental price divided by peak FP8 throughput.
Where does this NVIDIA GeForce RTX 5090 pricing data come from?
Prices are read daily from each provider's public price list or API and recorded whenever they change. Listings are shown per GPU-hour on-demand unless marked otherwise, and the chart covers the last 30 days. The newest listing was checked just now.