Novita AI Spot GPU Instances allow you to access spare GPU compute capacity at significantly reduced rates compared to On-Demand pricing.
Get RTX 4090 GPUs for as low as $0.18/hr (versus $0.35/hr On-Demand) or High-Frequency RTX 4090 GPUs at $0.35/hr (versus $0.69/hr On-Demand) — up to 50% savings on the same powerful hardware, making Spot Instances an ideal choice for fault-tolerant and flexible applications.
Whether you’re training machine learning models, running batch inference jobs, or processing large datasets, Spot Instances provide the same high-performance GPU capabilities at a fraction of the cost.
How Spot Instances Work
Spot Instances utilize Novita AI’s spare GPU capacity. This capacity is made available at lower prices because it can be reclaimed when demand for regular instances increases. By using Spot Instances, you can significantly reduce your GPU compute costs for appropriate workloads.
Key Characteristics
- Variable availability: Spot Instances may be interrupted when Novita AI requires the capacity back
- Significant cost savings: Access the same GPU performance at up to 50% less than On-Demand prices
- Protection period: Each Spot Instance includes a 1-hour protection window after launch
- Advance notifications: Receive interruption notices 1 hour before reclamation, with an additional 5-minute warning
Pricing and Available GPU Types
Spot Instance pricing delivers substantial savings. With Spot Instances, RTX 4090 24GB GPUs are available for as low as $0.18/hr (compared to $0.35/hr On-Demand), while High-Frequency RTX 4090 variants cost just $0.35/hr (versus $0.69/hr On-Demand). The High-Frequency configuration accelerates AI image generation by up to 150% through optimized CPU-GPU coordination.
| GPU Type | Spot Price | On-Demand Price | Savings |
|---|---|---|---|
| RTX 4090 (24GB) | $0.18/hr | $0.35/hr | ~49% |
| RTX 4090 High-Frequency | $0.35/hr | $0.69/hr | ~49% |
Interruption Handling
Protection Window
Every Spot Instance is protected from interruption for the first hour after launch. During this protection period, your instance will not be reclaimed regardless of capacity demands.
Interruption Notices
When Novita AI needs to reclaim Spot Instance capacity, the system provides:
- Initial warning: 1-hour advance notice before interruption
- Final warning: 5-minute notice before instance termination
These notifications allow you to:
- Save work and checkpoint progress
- Back up data to persistent storage
- Gracefully shut down applications
- Migrate to another Spot or On-Demand instance
Best Practices
Ideal Use Cases
Spot Instances work best for:
- Batch processing and distributed training with checkpointing
- Development/testing environments
- Fault-tolerant or stateless applications
- Flexible-deadline workloads
Before You Start
Ensure your workload can:
- Handle interruptions and resume from checkpoints
- Either complete within the 1-hour protection window or tolerate potential interruptions
Launch Your First Spot Instance in 3 Steps
Step 1: Access Your Console
Log in to your Novita AI GPU Console
Step 2: Switch to Spot Billing
In the right sidebar under Filter, change Billing Method to “Spot” to see discounted prices
Step 3: Deploy
Select your GPU configuration and click “Deploy”
That’s it! Your Spot Instance will launch with:
- 1-hour protection period
- Up to 50% cost savings
- 1-hour advance interruption notice
Pro tip: Implement checkpointing in your application to handle potential interruptions gracefully.
Comparison with On-Demand Instances
| Feature | Spot Instances | On-Demand Instances |
|---|---|---|
| Pricing | Up to 50% less | Standard rates |
| Availability | Subject to capacity | Always available |
| Interruption risk | May be reclaimed with notice | No interruptions |
| Protection period | 1 hour after launch | Continuous |
| Use case | Flexible, fault-tolerant workloads | Critical, uninterruptible workloads |
By choosing Spot Instances for appropriate workloads, you can access the same powerful GPU resources while optimizing your compute costs.
Novita AI is an AI cloud platform that offers developers an easy way to deploy AI models using our simple API, while also providing the affordable and reliable GPU cloud for building and scaling.
Discover more from Novita
Subscribe to get the latest posts sent to your email.
