Baseten Pricing
Every Baseten plan with its monthly and annual price, what it costs for common team sizes, and a calculator to estimate your own monthly cost before you buy.
Compare Baseten alternatives by priceRead our Baseten reviewCompare AI Software pricing
Plans and prices
| Plan | Monthly billing | Annual billing (per month) | Per user or flat | Users | Free tier |
|---|---|---|---|---|---|
| Pay as you go | Contact sales | Contact sales | — | No limit listed | No |
| Production scale | Contact sales | Contact sales | — | No limit listed | No |
Prices last updated November 2026.
What each Baseten plan includes
Pay as you go
- GPU and CPU compute billed per minute
- Elastic autoscaling, including scale to zero
- Usage-based billing with no contract or capacity planning
- Access to Model APIs, Dedicated Inference, Training and Sandboxes
Production scale
- Custom regions
- Custom support
- Volume discounts
- Capacity planning
- SLAs and custom terms
- Custom pricing via sales
- Reserved and priority compute
What Baseten costs by team size
| Team size | 1 user | 5 users | 10 users | 25 users | 50 users |
|---|---|---|---|---|---|
| Monthly cost | Contact for pricing | Contact for pricing | Contact for pricing | Contact for pricing | Contact for pricing |
The plan that fits each team size, billed monthly, before add-ons and usage charges.
Usage pricing
| Metric | Plan | Included per month | Price after that |
|---|---|---|---|
| A100 GPU (80 GiB) dedicated inference or training | Pay as you go | None | $0.0667 per minute |
| A10G GPU (24 GiB) dedicated inference or training | Pay as you go | None | $0.0201 per minute |
| B200 GPU (180 GiB) dedicated inference or training | Pay as you go | None | $0.1663 per minute |
| CPU instance 1 vCPU 2 GiB | Pay as you go | None | $0.0006 per minute |
| CPU instance 16 vCPU 64 GiB | Pay as you go | None | $0.0138 per minute |
| CPU instance 4 vCPU 16 GiB | Pay as you go | None | $0.0035 per minute |
| GPT OSS 120B Model API input tokens | Pay as you go | None | $0.10 per 1M tokens |
| GPT OSS 120B Model API output tokens | Pay as you go | None | $0.50 per 1M tokens |
| H100 GPU (80 GiB) dedicated inference or training | Pay as you go | None | $0.1083 per minute |
| H100 MIG GPU (40 GiB) dedicated inference or training | Pay as you go | None | $0.0625 per minute |
| L4 GPU (24 GiB) dedicated inference or training | Pay as you go | None | $0.0141 per minute |
| Sandbox snapshots storage | Pay as you go | None | $0.20 per GB-month |
| Sandbox volumes storage | Pay as you go | None | $0.16 per GB-month |
| Sandboxes CPU | Pay as you go | None | $0.0655 per vCPU-hour |
| Sandboxes memory | Pay as you go | None | $0.0211 per GB RAM-hour |
| T4 GPU (16 GiB) dedicated inference or training | Pay as you go | None | $0.0105 per minute |
Baseten Pricing Calculator
Estimate your team's real monthly and annual cost based on seats, billing cycle, and add-ons.
Cost breakdown
Estimated from the vendor's published prices, last updated November 2026. Actual costs may vary based on negotiated contracts. How we collect prices
Ready to try Baseten?
Use this estimate to align with your team, then start a trial or subscription.
Visit BasetenBetter value alternatives
See what similar tools would cost for the same team size.
Baseten pricing FAQs
No, Baseten does not list a free plan, and its plans are priced on request.
Yes. Baseten charges by usage for A100 GPU (80 GiB) dedicated inference or training, A10G GPU (24 GiB) dedicated inference or training, B200 GPU (180 GiB) dedicated inference or training and 13 more, past what the plan includes.