GPU cloud pricing
NVIDIA B300 price per hour: cheapest cloud
9 clouds rent out the B300. Hyperstack is cheapest at $7.40 per GPU-hour; Latitude.sh charges 2.2× as much.
- On-demand from$7.40per GPU-hour · Hyperstack
- One GPU, all month$5,402.00730 hours on-demand
- Spot from$6.27per GPU-hour · Verda (DataCrunch)
- Price gap2.2×across 9 clouds
- Memory288 GBlargest version listed
On-demand list prices per GPU-hour in USD · checked Nov 17, 2026 · How we compare GPU prices
Ranked for your usage
Monthly B300 cost by cloud
Cheapest for your volume
B300 at Hyperstack
9 options · most expensive $11,520.00, 2.2× this price
| Cloud | Estimated / month | Per GPU-hour | Spot / GPU-hour |
|---|---|---|---|
1 HyperstackGPU cloud · B300Pricing pageChecked | $5,328.00 | $7.40 | not listed |
2 ScalewayGPU cloud · B300Pricing pageChecked | $5,968.08 | $8.29 | not listed |
3 CrusoeGPU cloud · B300Pricing pageChecked | $6,451.20 | $8.96 | not listed |
4 RunPod (Secure Cloud)GPU cloud · B300Pricing pageChecked | $6,472.80 | $8.99 | not listed |
5 NebiusGPU cloud · B300Pricing pageChecked | $6,840.00 | $9.50 | not listed |
6 DigitalOceanGPU cloud · B300 · gpu-b300x1-288gbPricing pageChecked | $8,056.80 | $11.19 | not listed |
7 Verda (DataCrunch)GPU cloud · B300Pricing pageChecked | $9,021.60 | $12.53 | $6.27 |
8 Oracle CloudHyperscaler · B300 · BM.GPU.B300.8; bare metal, 8 GPUs; billed per second after tPricing pageChecked | $10,800.00 | $15.00 | not listed |
9 Latitude.shGPU cloud · B300 · g4-b300-largePricing pageChecked | $11,520.00 | $16.00 | not listed |
1 GPU × 24 h × 30 days = 720 GPU-hours at list price. Storage, network egress, taxes and reserved-capacity discounts are not included.
On-demand and spot list prices per GPU-hour in USD, checked daily (last Nov 17, 2026). Sources: GPU Rental Prices and Cloud GPU List Prices (AWS, Google Cloud and Azure price APIs), both CC BY 4.0; every row links the cloud's own pricing page. Order depends on price only. How we compare GPU prices.
Serverless B300: pay only while it runs
Billed per second of use, usually with a cold start; cheaper than renting for spiky traffic, dearer around the clock.
| Platform | Per GPU-hour of use | Source |
|---|---|---|
| Modal B300 | $7.10 | Pricing page |
| RunPod (Secure Cloud) B300 | $11.99 | Pricing page |
| fal B300 | $12.99 | Pricing page |
Before you rent
Rates are per GPU-hour; an instance with several GPUs bills all of them. Hyperscalers often sell this chip only in 8-GPU instances, so check the smallest size before comparing.
Running one B300 around the clock costs $5,402.00 a month at Hyperstack and $11,680.00 at Latitude.sh. Availability, regions and network speed differ, so check capacity before you commit.
Serving an open model? Compare per-token API prices first: below a steady load, paying per token is often cheaper than a rented GPU.