Baseten pricing 2026: plans and costs
Quick pricing summary
- Basic: $0 month, pay as you go
- Pro: Custom (custom quote)
- Enterprise: Custom (custom quote)
- Worth noting: Pay only for active compute, billed per minute; no charge for idle time. New accounts receive free credits. Compute discounts negotiable on Pro and Enterprise.
Plans
Basic
$0 month, pay as you go
- Dedicated deployments
- Model APIs
- Training
- Fast cold starts
- SOC 2 Type II and HIPAA compliant
- Email and in-app chat support
- Deployment options: Baseten
Pro
Custom custom quote
- Everything in Basic
- Unlimited autoscaling
- Priority access to high-demand GPUs
- Dedicated compute
- Higher Model API rate limits
- Hands-on engineering expertise
- Dedicated support on Slack and Zoom
- Volume discounts available
Enterprise
Custom custom quote
- Everything in Pro
- Custom SLAs
- Self-host deployments
- On-demand flex compute
- Use existing cloud commitments
- Full control over data residency
- Advanced security and compliance
- Custom global regions
- Advanced RBAC with Teams
- Deployment options: Baseten, Your VPC, Hybrid
- Volume discounts available
What you get on each plan
| Feature | Basic | Pro | Enterprise |
|---|---|---|---|
| Dedicated deployments | ✓ | ✓ | ✓ |
| Model APIs | ✓ | Higher rate limits | Higher rate limits |
| Training | ✓ | ✓ | ✓ |
| Priority access to high-demand GPUs | — | ✓ | ✓ |
| Dedicated compute | — | ✓ | ✓ |
| Support | Email and in-app chat | Dedicated Slack and Zoom | Dedicated Slack and Zoom |
| Custom SLAs | — | — | ✓ |
| Self-host deployments | — | — | ✓ |
| Advanced RBAC with Teams | — | — | ✓ |
| Deployment options | Baseten | Baseten | Baseten, Your VPC, Hybrid |
| Volume discounts | — | ✓ | ✓ |
Hidden costs & add-ons
| Cost driver | Price | When it applies |
|---|---|---|
| T4 16 GiB GPU | $0.01052/minute | Dedicated deployments and training |
| L4 24 GiB GPU | $0.01414/minute | Dedicated deployments and training |
| A10G 24 GiB GPU | $0.02012/minute | Dedicated deployments and training |
| A100 80 GiB GPU | $0.06667/minute | Dedicated deployments and training |
| H100 MIG 40 GiB GPU | $0.0625/minute | Dedicated deployments and training |
| H100 80 GiB GPU | $0.10833/minute | Dedicated deployments and training |
| B200 180 GiB GPU | $0.16633/minute | Dedicated deployments and training |
| CPU 1x2 | $0.00058/minute | Dedicated deployments |
| CPU 1x4 | $0.00086/minute | Dedicated deployments |
| CPU 2x8 | $0.00173/minute | Dedicated deployments |
| CPU 4x16 | $0.00346/minute | Dedicated deployments |
| CPU 8x32 | $0.00691/minute | Dedicated deployments |
Baseten pricing vs alternatives
| Brand | Entry price | Free plan | Model | Best for | |
|---|---|---|---|---|---|
| Baseten | Contact sales | Yes | Usage-based | — | |
| GitHub | Free; paid from $4 per seat / mo | Yes | Per seat | Development teams, from individuals to enterprises, that build, review, and ship software on one platform. | Compare → |
| MLflow | — | — | — | Teams of all sizes building LLM applications, agents and ML models who want open source tooling without vendor lock-in. | Compare → |
| Databricks | — | — | — | Data engineering, data science, analytics and AI teams at startups and global enterprises | Compare → |
| Amazon SageMaker AI | — | — | — | — | Compare → |
List prices as published on www.baseten.co/pricing/. Totals are worked examples, not quotes.