Loading prices…
RTX 4090D rental prices
NVIDIA · Ada Lovelace · 24 GB GDDR6X · PCIe · 2023
1 providers · 2 configurations · prices as of Nov 9, 2026, 23:02 UTC
Other versions:24GB
Lowest on-demand
$12.9per GPU·h · ≈ $9,434/month
at Vast.aiMedian on-demand
$12.9/GPU·h
across 1 providers
Lowest spot
$0.65/GPU·h
interruptible
FP16/BF16 cost
$87.9/PFLOP·h
#73 of 73 GPUs for value
Providers
1
2 configurations
Price history
Daily RTX 4090D prices per GPU-hour. The shaded period is reconstructed from public archives: the version history of SkyPilot's price catalogs, Internet Archive copies of providers' pricing pages and the open gpu-price-feed project (fewer providers than live data). After that, PetaPrice's own tracking.
Lowest on-demandMedian on-demandLowest spot
Price by provider
Cheapest On-demand RTX 4090D offer at each provider, per GPU-hour. Shorter and greener is cheaper.
Vast.ai🇺🇸$12.9best
Thin vertical line = median across providers ($12.9).
Every provider we track
Pricing
Size
Host
Provider HQ
1 providers · 2 configurations
Sort
- Vast.aicommunityOn-demandSpot🇺🇸 HQ in United Statesvia Vast.aifrom · On-demandCheapest$12.9/GPU·hCheapest
What can the RTX 4090D run?
Memory for the weights plus one 32K-token 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 | $/mo |
|---|---|---|---|---|---|
gpt-oss-20b OpenAI · 20.9B | 20.9B · 3.6B active | 20 GiB | 1 | $12.92 at Vast.ai | $9,434 |
Gemma 3 27B Google · 27.4B | 27.4B | 28 GiB | 2 | $25.85 at Vast.ai | $18,868 |
Qwen3-32B Alibaba · 32.8B | 32.8B | 39 GiB | 2 | $25.85 at Vast.ai | $18,868 |
Llama 3.3 70B Instruct Meta · 70.6B | 70.6B | 76 GiB | 4 | $51.69 at Vast.ai | $37,735 |
gpt-oss-120b OpenAI · 117B | 117B · 5.1B active | 110 GiB | 8 | $103.38 at Vast.ai | $75,470 |
Qwen3-235B-A22B Alibaba · 235B | 235B · 22B active | 225 GiB | 16 | $206.77 at Vast.ai | $150,941 |
DeepSeek-R1 DeepSeek · 684B | 684B · 37B active | 640 GiB | 32 | $413.54 at Vast.ai | $301,881 |
Kimi K3 Moonshot AI · 2.78T | 2.78T · 104B active | 2.53 TiB | 64+ | — | — |
Assumes 90% of each GPU's memory is usable and rounds up to 1, 2, 4, 8… GPUs for tensor parallelism (up to 64). $/hr is the GPU count times the cheapest in-stock on-demand rate per GPU; check that the provider rents instances that large. Serving many requests at once needs more memory for context.
Alternatives to the RTX 4090D
GPUs with similar compute and memory, and their lowest on-demand price right now.
- RTX 409024 GB · 0.165 PF FP16$0.34/h$2.06/PFLOP·h
- Titan RTX24 GB · 0.131 PF FP16$0.13/h$1.03/PFLOP·h
- RTX PRO 4000 Blackwell24 GB · 0.147 PF FP16$0.27/h$1.84/PFLOP·h
- RTX 508016 GB · 0.113 PF FP16$0.22/h$1.92/PFLOP·h
- RTX 509032 GB · 0.209 PF FP16$0.46/h$2.21/PFLOP·h
- A3024 GB · 0.165 PF FP16$0.32/h$1.97/PFLOP·h
RTX 4090D pricing questions
How much does it cost to rent an RTX 4090D?
On-demand RTX 4090D prices start at $12.9 per GPU-hour (Vast.ai). The median across 1 providers is $12.9, and the most expensive is $12.9. Running one GPU around the clock costs about $9,434 per month at the lowest price.
Is there cheaper spot or interruptible RTX 4090D capacity?
Yes. Spot (interruptible) RTX 4090D capacity starts at $0.65 per GPU-hour, 95% below the cheapest on-demand price. Spot instances can be preempted at any time, so use them for checkpointed training or batch jobs.
How much compute do you get per dollar with the RTX 4090D?
At $12.9 per hour and 0.147 PF of dense FP16/BF16, the RTX 4090D costs $87.9 per PFLOP-hour. That ranks 73 of 73 GPUs with on-demand prices on PetaPrice. Figures are datasheet peaks; real utilization is typically 30–60%.
What are the RTX 4090D's key specs?
24 GB GDDR6X with 1,008 GB/s of memory bandwidth, Ada Lovelace architecture (2023), PCIe form factor, 425 W. Dense peak: 0.147 PF FP16 / BF16, 0.294 PF FP8, 0.588 PF INT8, 0.0735 PF TF32, 0.0735 PF FP32, 0.0011 PF FP64.
How current are these prices?
Prices were collected on Nov 9, 2026, 23:02 UTC from provider APIs and public catalogs and refresh about every 2 hours. They are indicative list prices; always confirm on the provider's website before booking.