Preemptible VMs with Spot pricing
Access GPU capacity at market-driven prices, with more choice in how you run and spend.
Unlock GPU capacity for flexible workloads
Move work that can wait or restart into lower-price windows, such as experiments, queued jobs and other workloads with flexible timing.
Plan with transparent, market-driven pricing
Benefit from rates that respond to demand and available capacity to decide when and where to run.
Choose the price that works for your requirement
Set a maximum price per GPU-hour or follow the current spot price, choosing the approach that fits your workload’s priorities.
Two ways to run
Set your maximum price
For workloads where price is the priority
- Define the highest price per GPU-hour you’re willing to pay.
- Run when the spot price is at or below your limit and capacity is available, paying the applicable spot price.
- Your VM stops if the market price exceeds your limit.
Follow the market price
To avoid price-triggered interruptions
- Accept the current spot price as it changes.
- Avoid stops triggered by a maximum-price threshold.
- Your VM will only stop when capacity is needed elsewhere.
Note that both options use preemptible capacity. Setting a higher maximum or following the price does not reserve GPUs or guarantee continuous runtime.
Built for work that can pause and resume
Experiment and evaluate
Test model configurations, compare approaches and run evaluations without requiring a fixed completion time.
Process data
Run queued preprocessing, transformation and other GPU jobs that can retry or resume after interruption.
Fine-tune with checkpoints
Use spot capacity for shorter fine-tuning jobs or workflows designed to save progress and recover.
Run batch inference
Process offline inference jobs that can be retried and do not need an immediate response.
Access now in three steps
Access now in three steps
-
Choose your GPU and region where capacity is available
-
Set your maximum price or follow the market
-
Launch your workload
Live Spot Preemptible VM prices
Access market-driven prices across GPU platforms and regions.
Get started
Access GPU capacity at market-driven prices, with complete control over what you’re willing to pay.
Frequently asked questions
Yes. All Preemptible Virtual Machines on the Nebius AI Cloud will with dynamic spot pricing. You’ll see “preemptible” in the console and documentation.