Builder
Prototypes & side projects
$0forever
no card required
Included tokens1M / mo
- 5 open-weight models
- 3 requests / second
- 1 project workspace
- Community Discord support
Scale your AI agents, not your bill.
Every plan runs on the same low-latency inference grid. Pay for the intelligence you use — and upgrade only when your agents do.
Scale
Teams shipping to production
$49/mo
billed monthly
Included tokens50M / mo
- All 40+ models + fine-tuning
- 100 requests / second
- Agent memory & vector store
- Usage analytics dashboard
- Priority email + chat support
Enterprise
Organizations at scale
Custom
annual contract
Included tokensUnlimited
- Dedicated GPU clusters
- SSO · SOC 2 · HIPAA
- 99.99% uptime SLA
- Named solutions engineer
Prices in USD. Every plan includes streaming + batch APIs, region pinning, and no cold-start fees — cancel anytime.