Pricing Overview
PureAI uses a credit-based billing system. Purchase credits in USD and use them to deploy and run models on GPU instances.
How It Works
- Purchase Credits: Add credits to your account via Stripe
- Deploy Models: Select a model and instance type
- Pay Per Hour: Credits are charged per hour of active deployment
- Monitor Usage: Track costs in the dashboard
Pricing Model
| Component | Pricing |
|---|
| Credits | 1 credit = $1 USD |
| Deployments | Per-hour based on instance type |
| API Calls | Included in deployment cost |
Credit Purchase Options
| Amount | Credits |
|---|
| $5 | 500 credits |
| $10 | 1,000 credits |
| $20 | 2,000 credits |
| $50 | 5,000 credits |
| $100 | 10,000 credits |
| Custom | Minimum $1 |
Cost Example
Deploying LLaMA 3.1 8B on GPU XS:
| Component | Cost |
|---|
| Instance | ~$0.20/hour |
| 8 hours of usage | ~$1.60 |
| 24 hours of usage | ~$4.80 |
Instance Tiers
| Tier | Starting Price | Models |
|---|
| XS | ~$0.20/h | 7B - 13B |
| S | ~$0.60/h | 13B - 34B |
| M | ~$1.80/h | 30B - 70B INT4 |
| L | ~$3.50/h | 70B FP16 |
| XL | ~$12.00/h | 70B - 180B |
| XXL | ~$20.00/h | 405B |
Prices are spot prices and may vary based on availability. On-demand pricing is higher.
Cost Management
- Auto-pause: Deployments can be paused when not in use
- Alerts: Set up notifications for usage thresholds
- Dashboard: Real-time cost tracking and projections
Next Steps
