Skip to content
InfratoApp
GPU prices / RTX-4090-24GB

NVIDIA RTX 4090 24 GB

RTX-4090-24GB costs $0.16 to $0.92 per GPU-hour on demand across 13 cloud providers, cheapest at Salad. Prices were read from each provider's own pricing page, most recently on 2026-10-27.
Cheapest read 26 Oct · spread 5.8× · a missing provider is never an estimate
CHEAPEST $0.16/hrMEDIAN $0.46/hrSPREAD 5.8×PROVIDERS 13

Price by provider — cheapest first

Provider$/GPU·hr ↓Δ 7dTermsRead
Salad$0.16— 0.0%not stated26 Oct ↗
Theta EdgeCloud$0.26— 0.0%not stated26 Oct ↗
TensorDock$0.35— 0.0%not stated26 Oct ↗
Packet·ai$0.39— 0.0%on demand27 Oct ↗
Shadeform$0.40— 0.0%on demand26 Oct ↗
Beam$0.44— 0.0%not stated27 Oct ↗
GPUhub$0.46▲ +4.5%not stated26 Oct ↗
Vast.ai$0.47▲ +6.8%on demand27 Oct ↗
Oblivus$0.64— 0.0%on demand26 Oct ↗
Lium$0.65▲ +1.6%not stated27 Oct ↗JSON
Spheron$0.72▲ +50.0%not stated26 Oct ↗
RunPod$0.89— 0.0%on demand27 Oct ↗
Runcrate$0.92▲ +76.9%not stated26 Oct ↗
Why a provider might be missing: its price is behind a quote form, older than 21 days, or flagged for review. A missing row is never an estimate.

Price history — cheapest daily reading

⚑ 7 of 12 weeks measuredWEEKLY LOW SHOWN
2026-08-04A FLAT MARK IS A WEEK WE DID NOT READ2026-10-20

Cost calculator

GPUs1×
Hours / month730 h
730 = 24/7 · on-demand rates only
Salad$117 /mo
Theta EdgeCloud$190 /mo
TensorDock$255 /mo
Packet·ai$285 /mo
Shadeform$292 /mo
Beam$321 /mo
GPUhub$336 /mo
Vast.ai$343 /mo
Oblivus$467 /mo
Lium$475 /mo
Spheron$526 /mo
RunPod$650 /mo
Runcrate$672 /mo

Spec sheet

VRAM24 GB
Bandwidth1.008 TB/ssource
FP16165 TFLOPSsource
FP8330 TFLOPSsource
TDP450 Wsource
NVLinkPCIe 4 (no NVLink)source
ArchitectureAda Lovelace
Released2022source
8 of 8 figures held

Good for

Local dev/prototyping & single-GPU inference of <=8B
reasoned from the specification, not measured · source

Benchmarks

resnet50 Offline
87,946 Samples/s
v6.0 · 2 accelerators · GATEOverflow
submission ↗
retinanet Offline
1,760 Samples/s
v5.1 · 2 accelerators · GATEOverflow
submission ↗
retinanet Server
1,449 Queries/s
v5.1 · 2 accelerators · GATEOverflow
submission ↗
retinanet MultiStream
226 Latency (ms)
v5.1 · 1 accelerator · GATEOverflow
submission ↗
whisper Offline
136 Samples/s
v5.1 · 1 accelerator · GATEOverflow
submission ↗
retinanet SingleStream
22.7 Latency (ms)
v5.1 · 1 accelerator · GATEOverflow
submission ↗

LLMs that fit this GPU’s VRAM

FP16
~9B params
FP8
~19B params
FP4/INT4
~38B params
Mistral 7B @FP16; Llama 3.1 8B @FP8; 13B @INT4
ceilings at each precision, reasoned from VRAM · not a benchmark

Frequently asked

How much does an RTX-4090-24GB cost per hour right now?
$0.16 an hour at Salad, the cheapest of 13 sellers we read. The dearest posts $0.92 at Runcrate. Every rate links to the page it was read from, with the date it was read.
Why do RTX-4090-24GB prices vary 5.8× between providers?
Because a GPU-hour is not one product. $0.16 at Salad and $0.92 at Runcrate buy different things: interconnect, host CPU and RAM, storage, egress, support, and whether the capacity is dedicated or shared. We publish the posted rate for each and the page it came from, and do not blend them into one number.
What is the cheapest way to rent an RTX-4090-24GB long-term?
We hold no committed or reserved rate for this card — every figure here is an on-demand posted rate. Committed pricing is usually quoted rather than published, so the honest answer is that we cannot tell you, rather than a guess at a discount.
How fresh are these prices?
These rates were read between 2026-10-26 and 2026-10-27. Nothing older than 21 days is shown: a reading that goes stale is hidden rather than displayed with an old date, because a wrong price is worse than a missing one. Each row links to the page its figure came from, so any of them can be checked against the seller.
Which is cheaper for an RTX-4090-24GB — a region or a commitment?
Both move the number and they move it independently. Where a provider prices the same card in several regions we show the cheapest they charge anywhere and name the region, because a rate you can only get in Estonia is still a rate you can get. A committed term is a different contract rather than a discount on this one, so committed rates are kept out of the headline and shown separately.
Can a provider pay to be listed first for the RTX-4090-24GB?
No. The table has one stated sort at a time and nothing in it can be bought — a sponsor cannot enter it, reorder a row or change a rate. How the site is paid for, and the line each source of money does not cross, is set out under money and independence at /our-data#sec5.
Vs its class
Live cheapest rates
GPUCheapest $/hrVRAMFP8 TFLOPS
RTX-3090-24GB$0.09○○
RTX-A4000-16GB$0.10○○
RTX-4090-24GB ← this$0.1624 GB○
RTX-5090-32GB$0.25○○
RTX-A4500-20GB$0.25○○
RTX-A5000-24GB$0.27○○

Cite this page

infratoapp, “RTX-4090-24GB cloud pricing”, read 26 October 2026. $0.16 per GPU-hour at Salad. https://www.infratoapp.com/gpu/RTX-4090-24GB
Reading date travels inside the citation. How we read this data →