Try
10 GPU-hours
$2.80
$0.28/hour
- Test your setup end to end
- Same GPU and image as a long run
- Unused hours never expire
Quick answer
How much does it cost to rent an RTX 4090 per hour?
As of 29 October 2026, renting an on-demand NVIDIA RTX 4090 costs $0.33 per hour at Salad and $0.89 at RunPod. They are the only two providers with a current self-serve on-demand price. Salad's price rises to $0.44 on 1 November. On GPUwithCrypto an RTX 4090 costs $0.28/hour, billed per minute and paid in crypto with no KYC.
How much does an RTX 4090 cost per hour at each provider?
According to GPUwithCrypto's price check of 29 October 2026, only 2 of the 24 providers we priced list a current self-serve on-demand RTX 4090. RunPod's Secure Cloud price is 2.7 times Salad's High Priority price. Marketplace and interruptible offers fill the rest.
| Provider | Configuration | On-demand $/GPU-hour | Spot / marketplace $/GPU-hour | Billing increment | Accepts crypto | ID / KYC |
|---|---|---|---|---|---|---|
| GPUwithCrypto | RTX 4090 24GB, whole GPUs | $0.28 | — (on-demand only) | Per minute | Yes | No |
| Salad | RTX 4090, High Priority tier, distributed consumer GPUs | $0.33 until 31 Oct; $0.44 from 1 Nov | $0.16 (Batch priority, most interruptible) | Per second | Yes (USDC, USDT, RENDER on Solana) | None documented |
| RunPod | RTX 4090, 1 GPU, Secure Cloud | $0.89 | $0.34 (Community Cloud) | Per second | Yes | KYC before first crypto payment |
| Spheron | RTX 4090 | Excluded: $0.72 marked 'last known Sep 17' | — | Per minute, 20-minute minimum | Yes (USDC) | None documented |
| Vast.ai | RTX 4090, marketplace | — | $0.13 (Vast.ai's public price feed, cheapest rentable offer) | Per second | Yes (BitPay, Crypto.com) | None documented |
| Akash Network | RTX 4090, decentralized marketplace | — | $0.34 (recent provider bids) | — | Yes (AKT, ACT) | None documented |
| TensorDock | RTX 4090, marketplace | — | $0.35 'typical' (page dated July 24, 2024; not counted) | Per second | Unknown | None documented |
| Shadeform | RTX 4090, 8 GPUs, resold from other clouds | — | None available | — | Yes (USDC, per Stripe) | None documented |
Prices checked on
USD per GPU-hour, taxes excluded. Salad's figures come from its official price-change post of 26 October 2026; salad.com/pricing already shows the 1 November prices. RunPod's Community Cloud is community-hosted. 'None documented' means no ID step in the official pages we read.
What changes on 1 November 2026?
Salad's High Priority RTX 4090 goes from $0.33 to $0.44 per hour, a 33% rise announced in its price-change post of 26 October 2026. High Priority is Salad's tier that is never preempted, and it set our reference price.
Our price was computed from the $0.33 figure, so it stays below Salad's new rate. Salad's cheaper tiers remain: from 1 November, Medium costs $0.347, Low $0.253 and Batch $0.160. Those tiers can be preempted.
Why are marketplace RTX 4090s cheaper than ours?
Because most of them run on independent hosts' machines, at prices the hosts set, and many can be interrupted. On 29 October 2026, Vast.ai's public price feed listed 562 rentable RTX 4090 offers, from $0.13 with a median of $0.49. Salad's Batch tier, its most interruptible, cost $0.16.
Both were below our $0.28. They suit image batches and LoRA runs that can restart. For a ComfyUI server or a training run that must not stop, compare on-demand prices. Instances are on-demand, not spot: nobody can take your GPU back while your balance covers it, and the instance runs until you stop it. The spot GPU guide explains the trade-off.
How much does an RTX 4090 cost per day, week or month?
Our price for one GPU next to Salad's High Priority price before and after 1 November 2026, and the median of 29 October.
| Period | GPU-hours | GPUwithCrypto | Salad until 31 Oct ($0.33) | Salad from 1 Nov ($0.44) | Median elsewhere ($0.61) |
|---|---|---|---|---|---|
| 1 hour | 1 | $0.28 | $0.33 | $0.44 | $0.61 |
| 1 day | 24 | $6.72 | $7.92 | $10.56 | $14.64 |
| 1 week | 168 | $47.04 | $55.44 | $73.92 | $102.48 |
| 30 days | 720 | $201.60 | $237.60 | $316.80 | $439.20 |
Prices checked on
Our day, week and month match the 24, 168 and 720 GPU-hour blocks, and unused GPU-hours never expire. Salad credits expire 12 months after purchase, per its billing docs.
What are the RTX 4090's specs?
| Architecture | Ada Lovelace, PCIe Gen4 desktop card |
|---|---|
| Memory | 24 GB GDDR6X |
| Memory bandwidth | 1,008 GB/s |
| FP8 Tensor (FP32 accumulate) | 330.3 TFLOPS dense, 660.6 with sparsity |
| FP16 / BF16 Tensor (FP32 accumulate) | 165.2 TFLOPS dense, 330.4 with sparsity |
| FP32 | 82.6 TFLOPS |
| NVLink | No |
| Total graphics power | 450 W |
| Marketing 'AI TOPS' | 1,321: sparse FP8 with FP16 accumulate, not comparable with dense data-center figures |
What fits on one RTX 4090?
Memory needed for common open models, using weights × 1.2 for inference (EleutherAI's rule of thumb). Image pipelines also load text encoders and a VAE. The VRAM guide has the formulas.
| Model and job | Memory needed | One RTX 4090 24GB? |
|---|---|---|
| Llama 3.1 8B inference, BF16 | ≈ 19.2 GB | ✓ |
| Mistral 7B inference, BF16 | ≈ 17.5 GB | ✓ |
| Llama 3.1 8B LoRA fine-tune | ≈ 19.2 GB | ✓ |
| Llama 3.1 8B QLoRA fine-tune | ≈ 6 GB | ✓ |
| SDXL 1.0 base, FP16 weights | 7 GB | ✓ |
| FLUX.1 [dev] transformer, FP8 weights | 12 GB, plus text encoders and VAE | ✓ |
| FLUX.1 [dev] transformer, BF16 weights | 24 GB | Tight: the weights alone fill the card |
| FLUX.1 [dev] LoRA training, BF16 base | ≈ 28.8 GB | No; QLoRA (≈ 9 GB) fits |
| Llama 3.1 70B inference, 4-bit | ≈ 42 GB | ✗ |
What does an RTX 4090 cost on GPUwithCrypto?
$0.28/hour, at least 15% under its named on-demand reference: Salad at $0.33, read on 29 October 2026. We do not claim to be the cheapest for every GPU; spot and marketplace offers (Vast.ai) can be lower. Instances are billed per minute at the hourly rate while they run, from the GPU-hours you bought for that GPU. Stopping an instance stops the charge at once. GPU-hours are sold in blocks of 10, 24, 168 or 720 hours for one GPU model, at the same hourly rate; instances of that GPU draw from them per minute, and unused hours never expire.
Popular
Day
24 GPU-hours
$6.72
$0.28/hour
- One day of continuous running
- A fine-tuning run or a batch job
- Unused hours never expire
Week
168 GPU-hours
$47.04
$0.28/hour
- One week of continuous running
- Training runs and evaluations
- Unused hours never expire
Month
720 GPU-hours
$201.60
$0.28/hour
- 30 days of continuous running
- An inference server running all month
- Unused hours never expire
In every plan: Whole GPUs, never shared · Root SSH, CUDA and PyTorch ready · Billed per minute · Unused hours never expire
RTX 4090 rental questions
What is the cheapest RTX 4090 cloud?
Among on-demand RTX 4090s that are never preempted, the lowest list price on 29 October 2026 was Salad's High Priority tier at $0.33, which becomes $0.44 on 1 November. GPUwithCrypto's $0.28/hour is at least 15% under that reference. Marketplace offers go lower: Vast.ai's public price feed started at $0.13.
How much does an RTX 4090 cost per month?
720 GPU-hours cost $201.60 on GPUwithCrypto. At Salad's High Priority price from 1 November 2026 the same month costs $316.80, and $439.20 at the median of 29 October.
Is an RTX 4090 enough for Flux and ComfyUI?
For inference, yes with FP8 weights: FLUX.1 [dev] takes 12 GB in FP8 and 24 GB in BF16. LoRA training in BF16 needs about 28.8 GB, so use QLoRA or a 32 GB RTX 5090. See the best GPU for ComfyUI and Flux.
RTX 4090 or RTX 5090: which should I rent?
The RTX 5090 has 32 GB against 24 GB and 1,792 GB/s of bandwidth against 1,008 GB/s. On GPUwithCrypto it costs $0.42/hour against $0.28/hour. Pick it when 24 GB is not enough. Prices are on the RTX 5090 page.
Can I rent an RTX 4090 with crypto?
Yes. GPUwithCrypto takes BTC, USDT, USDC, ETH, SOL, LTC with no ID and no card. Salad takes USDC, USDT and RENDER on Solana, Vast.ai uses BitPay and Crypto.com, and RunPod asks for KYC before the first crypto payment.
Sources (13)
- SaladCloud price changes, November 2026 blog.salad.com/saladcloud-price-changes-november-2026 — Salad (read 2026-10-29)
- SaladCloud — Billing docs.salad.com/general/explanation/billing — Salad (read 2026-10-29)
- Runpod — Pricing runpod.io/pricing — RunPod (read 2026-10-29)
- Runpod — Billing information docs.runpod.io/references/billing-information — RunPod (read 2026-10-29)
- Spheron — Pricing spheron.network/pricing — Spheron (read 2026-10-29)
- Vast.ai public GPU pricing feed — Vast.ai (read 2026-10-29)
- Akash Network — GPU prices API console-api.akash.network/v1/gpu-prices — Akash Network (read 2026-10-29)
- TensorDock — Cloud GPUs tensordock.com/cloud-gpus.html — TensorDock (read 2026-10-29)
- Shadeform — Instance types API api.shadeform.ai/v1/instances/types — Shadeform (read 2026-10-29)
- NVIDIA GeForce RTX 4090 — NVIDIA (read 2026-10-29)
- NVIDIA Ada GPU architecture whitepaper — NVIDIA (read 2026-10-29)
- FLUX.1 [dev] model card — Black Forest Labs (read 2026-10-29)
- Transformer Math 101 — EleutherAI (read 2026-10-29)