How Much VRAM Do I Need for LLM Training and Inference?
A practical engineer's guide to VRAM for LLM training, fine-tuning, and inference. Learn how to estimate VRAM for 7B and 70B models, and what to...
2026-10-25 Read guideLive prices · updated daily
Find the cheapest cloud GPU to rent
Compare hourly on-demand and spot prices for H100, A100, L40S, RTX 4090 and more across Vast.ai and Runpod. Normalised pricing, 30-day price history and free cost calculators.
- 2
- Providers tracked
- 37
- Live listings
- 17
- GPU models
- 3
- Languages
Live GPU price comparison
Filter by provider, GPU or region — filters run locally in your browser.
| Provider | GPU model | VRAM | GPUs | vCPU | Region | On-demand /hr | Spot /hr | Actions |
|---|---|---|---|---|---|---|---|---|
| Vast.ai | RTX 3090 | 24 GB | 1 | — | Distributed (global hosts) | $0.0881 Best price | $0.0840 | Rent this GPU |
| Vast.ai | RTX 4090 | 24 GB | 1 | — | Distributed (global hosts) | $0.1081 Best price | $0.0533 | Rent this GPU |
| RunPod | RTX A5000 | 24 GB | 1 | — | Global (secure + community cloud) | $0.1600 Best price | — | Rent this GPU |
| Vast.ai | V100 | 16 GB | 1 | — | Distributed (global hosts) | $0.1622 Best price | $0.1333 | Rent this GPU |
| RunPod | RTX A4000 | 16 GB | 1 | — | Global (secure + community cloud) | $0.1700 Best price | — | Rent this GPU |
| Vast.ai | RTX 5090 | 32 GB | 1 | — | Distributed (global hosts) | $0.2129 Best price | $0.1000 | Rent this GPU |
| RunPod | RTX 3090 | 24 GB | 1 | — | Global (secure + community cloud) | $0.2200 | — | Rent this GPU |
| RunPod | V100 | 16 GB | 1 | — | Global (secure + community cloud) | $0.2300 | — | Rent this GPU |
| RunPod | RTX A6000 | 48 GB | 1 | — | Global (secure + community cloud) | $0.3300 Best price | — | Rent this GPU |
| Vast.ai | A100 PCIe | 40 GB | 1 | — | Distributed (global hosts) | $0.3370 Best price | — | Rent this GPU |
| RunPod | RTX 4090 | 24 GB | 1 | — | Global (secure + community cloud) | $0.3400 | — | Rent this GPU |
| RunPod | A40 | 48 GB | 1 | — | Global (secure + community cloud) | $0.3500 Best price | — | Rent this GPU |
| RunPod | L4 | 24 GB | 1 | — | Global (secure + community cloud) | $0.4400 Best price | — | Rent this GPU |
| Vast.ai | L40S | 48 GB | 1 | — | Distributed (global hosts) | $0.4689 Best price | $0.4670 | Rent this GPU |
| RunPod | H200 SXM | 141 GB | 1 | — | Global (secure + community cloud) | $3.5900 Best price | — | Rent this GPU |
| RunPod | RTX 5090 | 32 GB | 1 | — | Global (secure + community cloud) | $0.6900 | — | Rent this GPU |
| RunPod | RTX 6000 Ada | 48 GB | 1 | — | Global (secure + community cloud) | $0.7400 Best price | — | Rent this GPU |
| RunPod | L40S | 48 GB | 1 | — | Global (secure + community cloud) | $0.7900 | — | Rent this GPU |
| RunPod | A100 SXM | 80 GB | 1 | — | Global (secure + community cloud) | $1.3900 | — | Rent this GPU |
| Vast.ai | A100 SXM | 80 GB | 1 | — | Distributed (global hosts) | $1.0422 Best price | — | Rent this GPU |
| RunPod | A100 PCIe | 80 GB | 1 | — | Global (secure + community cloud) | $1.1900 | — | Rent this GPU |
| RunPod | H100 PCIe | 80 GB | 1 | — | Global (secure + community cloud) | $1.9900 Best price | — | Rent this GPU |
| Vast.ai | H100 SXM | 94 GB | 1 | — | Distributed (global hosts) | $2.4741 Best price | $2.2000 | Rent this GPU |
| RunPod | H100 SXM | 80 GB | 1 | — | Global (secure + community cloud) | $2.6900 | — | Rent this GPU |
| RunPod | B200 | 180 GB | 1 | — | Global (secure + community cloud) | $5.9800 Best price | — | Rent this GPU |
| Vast.ai | B200 | 180 GB | 1 | — | Distributed (global hosts) | $6.2521 | $6.2500 | Rent this GPU |
| Vast.ai | H200 SXM | 141 GB | 1 | — | Distributed (global hosts) | $14.8339 | $5.3333 | Rent this GPU |
No offers match these filters. Relax the price limit or clear the search.
Popular GPUs
Lowest current single-GPU rate for each model, with a 30-day trend.
RTX 3090
24 GB · 2 live offers
from $0.0881 /GPU/hr View detailsRTX 4090
24 GB · 3 live offers
from $0.1081 /GPU/hr View detailsRTX A5000
24 GB · 1 live offers
from $0.1600 /GPU/hr View detailsV100
16 GB · 2 live offers
from $0.1622 /GPU/hr View detailsRTX A4000
16 GB · 1 live offers
from $0.1700 /GPU/hr View detailsRTX 5090
32 GB · 2 live offers
from $0.2129 /GPU/hr View detailsRTX A6000
48 GB · 1 live offers
from $0.3300 /GPU/hr View detailsA100 PCIe
40 GB · 2 live offers
from $0.3370 /GPU/hr View detailsA40
48 GB · 2 live offers
from $0.3500 /GPU/hr View detailsL4
24 GB · 1 live offers
from $0.4400 /GPU/hr View detailsL40S
48 GB · 3 live offers
from $0.4689 /GPU/hr View detailsRTX 6000 Ada
48 GB · 1 live offers
from $0.7400 /GPU/hr View detailsFree calculators
Both calculators run entirely in your browser — the results stay in the URL, so you can share a configuration.
GPU cost estimator
Estimate the monthly cost of renting GPUs by the hour (on-demand vs spot), including storage and egress assumptions.
Instant results Open calculatorLLM VRAM calculator
Work out how much GPU memory a large language model needs for weights, KV cache and activations — and which GPUs can host it.
10 model presets Open calculatorWhy trust these numbers
Collected, never invented
A scheduled collector reads providers' own pricing sources daily and stores a dated snapshot. Source type is recorded per row.
Ranked by price, not by payout
Sorting and verdicts are computed from the data. Commissions never influence what you see, and links are never cloaked or redirected.
Works without JavaScript
Tables, FAQs, charts and calculator results are rendered on the server. Turn JS off and nothing breaks.
Comparisons
Side-by-side pricing with a computed verdict for each pairing.
Latest guides
Evergreen, vendor-neutral guides on renting cloud GPUs for LLM training, fine-tuning and inference.
Cloud GPU Cost Estimator: A Practical Guide for ML Engineers
Build a cloud GPU cost estimator that survives real invoices. I cover the inputs, the line items people forget, and how to sanity-check any rental quote.
2026-10-25 Read guideH100 vs A100 Rental Price Per Hour: What the Rate Gap Actually Buys You
Comparing H100 vs A100 rental price per hour from an ML engineer's view: raw compute, memory bandwidth, power, and where the premium is worth it.
2026-10-25 Read guideCloud GPU Rental Price Comparison: A Field Guide for Engineers
Compare cloud GPU rental prices per hour across providers. Learn the real cost drivers, hourly vs monthly pricing, and how to use a cloud GPU...
2026-10-25 Read guideH100 Rental Pricing: What You Actually Pay Per Hour
A practical h100 rental pricing guide for ML engineers. Compare h100 cloud pricing, rental rates per GPU, and costs for training vs inference.
2026-10-24 Read guideA100 GPU Rental: What I Actually Check Before Renting
Renting an A100 for LLM training, fine-tuning, or inference? Here's what I check before paying: multi-GPU scaling, memory limits, and price per hour.
2026-10-24 Read guideFrequently asked questions
Everything about renting cloud GPUs, affiliate tracking and how we collect pricing data.
Is this site a GPU marketplace?
No. We do not sell, resell or broker compute and we never take payment. We publish normalised pricing collected from providers' public pages and link you to the provider so you rent directly from them.
How often are the prices updated?
A scheduled collector runs every 6 hours (00:00, 06:00, 12:00 and 18:00 UTC). Every table shows the timestamp of the last successful collection. If a source fails, the run is logged and an operational alert is raised rather than silently reusing stale data.
How do you make money?
Through the Vast.ai referral programme. If you click a provider link and later pay for compute, the provider may pay us a commission. You never pay more, and commissions never change the prices, ordering or verdicts on this site.
What does on-demand and spot mean here?
On-demand (also called uninterruptible or secure) means the instance stays yours for as long as you pay. Spot, interruptible or community capacity can be reclaimed by the host and is therefore cheaper. Prices are normalised to a single GPU per hour so different configurations can be compared fairly.
Do I need an account to use the calculators?
No. Both calculators run entirely in your browser and the results are stored in the URL fragment, so you can bookmark or share a configuration. Nothing is sent to our servers.
Why do some providers show no price for a GPU?
Either the provider does not currently list that model, or it was listed but the offer was withdrawn before the last collection. We would rather show a gap than an invented number.
Volume buyers
B2B bulk GPU enquiry
Reserving large amounts of GPU capacity? Tell us what you need and we will forward your request to the right provider team.
- Forwarded by email only — nothing is stored on this site
- Typical first reply from a provider team: 1–2 business days
- Useful for reserved clusters, multi-month commitments and data-residency requirements