Vast.ai Pricing Calculator
Updated June 2026Marketplace usage (dynamic $/hr GPU + $/GB/hr storage + $/TB bandwidth) with reserved pre-pay discounts
Vast.ai pricing: Pricing breakdown for Vast.ai — the GPU rental marketplace with dynamic per-second pricing, interruptible spot bids, reserved discounts, and serverless. Use the free calculator below to enter your usage and get an instant all-in monthly estimate — including overages — and see which plan is cheapest for your needs.
How Vast.ai prices: Marketplace usage (dynamic $/hr GPU + $/GB/hr storage + $/TB bandwidth) with reserved pre-pay discounts.
How Vast.ai prices
Marketplace usage (dynamic $/hr GPU + $/GB/hr storage + $/TB bandwidth) with reserved pre-pay discounts
Your usage
💡 8h/day on one GPU ≈ 240 GPU-hrs/mo; running 24/7 ≈ 720 GPU-hrs/mo
💡 Multi-GPU hosts bill ~per-GPU rate × GPU count (e.g. 8× H200, 4× B200).
Monthly Estimate
Marketplace (pay-as-you-go)
$163.68
per month
Cost Breakdown
Annual
$1,964.16
Why Marketplace (pay-as-you-go) — $163.68/mo
- Marketplace (pay-as-you-go) at $163.68/mo.
- How we got there: 240 GPU-hrs (GPU-hours / month) × 1 $/GPU-hr (GPU type (on-demand market rate)) × 1 GPUs (GPUs per machine) = 164 $.
- GPU compute spend (163.68 $ × $1) is 100% of your bill — the lever that matters most.
- Need less? GPU type (on-demand market rate): RTX 5060 Ti 16GB ($0.194/hr) drops you to Marketplace (pay-as-you-go) at $46.56/mo.
Budget range
Plan for usage swings, not just today's estimate.
Conservative
$81.84
120 GPU-hrs/mo
Expected
$163.68
240 GPU-hrs/mo
Aggressive
$327.36
480 GPU-hrs/mo
Need a calculator like this on your pricing page?
Embed interactive pricing calculators on your website to help customers understand costs and boost conversions.
Get StartedAbout this Vast.ai calculator
This calculator estimates your Vast.ai cost from publicly available pricing. Actual costs may vary with your specific agreement, volume discounts, and usage patterns — always verify on the provider's official pricing page for the most current rates.
Vast.ai pricing — frequently asked questions
How much does a GPU cost on Vast.ai? ▼
There is no fixed price — Vast.ai is a marketplace where each host sets its own per-second rate, so prices float with supply and demand. As captured in June 2026, on-demand rates ran roughly $0.682-$0.875/hr for an RTX 4090, $1.166-$1.205/hr for an A100 80GB, $2.525/hr per H100 NVL, and $4.492/hr per B200, before bandwidth.
What is the difference between on-demand and interruptible instances? ▼
On-demand instances are guaranteed and run at the host's fixed rate. Interruptible instances are a spot/auction bid — they are typically 50%+ cheaper but can be paused (preempted) when another user outbids you, making them best for fault-tolerant, checkpointable workloads.
Can I get a discount for long-term GPU use on Vast.ai? ▼
Yes. You can convert any on-demand instance to a Reserved instance by pre-paying for a 1-, 3-, or 6-month term, for up to 50% off. Pre-paid credits are locked to that specific machine and cannot migrate between hosts.
How does Vast.ai bill for storage and bandwidth? ▼
Storage is charged per GB per hour for every second your instance exists (including stopped instances — only deleting the instance stops storage billing). Bandwidth is charged per TB for both upload and download, at host-set rates, on top of the GPU $/hr.
Is this Vast.ai pricing calculator free and accurate? ▼
Yes — it's completely free with no signup. It uses the latest publicly available Vast.ai pricing (verified June 2026). Actual costs may vary with volume discounts or enterprise terms — always confirm on Vast.ai's official pricing page.