Pricing designed for production usage
Try the full loop for free; pay for production quotas. Overage is billed at cost-plus rates — your live traffic is never cut off for hitting a quota.
Free
The full build-to-run loop, free
- 100 runs / month
- 500K LLM tokens / month
- 1 workspace · 1 API endpoint
- Basic traces (7-day retention)
Pro
Production for individual developers
- 2,000 runs / month
- 10M LLM tokens / month
- 3 workspaces · 10 API endpoints
- Full trace & replay (30 days)
Team
Collaboration and higher quotas
- 10,000 runs / month
- 50M LLM tokens / month
- 10 workspaces · 50 API endpoints
- All model tiers (incl. quality)
Enterprise
Private deployment & governance
- Unlimited workspaces & members
- SSO / SAML · 1-year+ audit logs
- Private deployment (Helm/K8s)
- 99.95% SLA · dedicated support
Full comparison
| Free | Pro | Team | Enterprise | |
|---|---|---|---|---|
| Quotas | ||||
| Workspaces | 1 | 3 | 10 | Unlimited |
| Members / workspace | 1 | 3 | 10 | Unlimited |
| Runs / month | 100 | 2,000 | 10,000 | Custom |
| LLM tokens / month | 500K | 10M | 50M | Custom |
| Knowledge base storage | 50MB | 5GB | 50GB | Custom |
| Published API endpoints | 1 | 10 | 50 | Unlimited |
| Capabilities | ||||
| Model tiers | fast only | fast + balanced | All (incl. quality) | All + BYOK |
| Trace / replay | Basic (7 days) | Full (30 days) | Full (90 days) | Full (configurable) |
| Audit log retention | 7 days | 30 days | 90 days | 1 year+ |
| SSO / SAML | — | — | — | Included |
| Private deployment | — | — | — | Included |
| Service | ||||
| SLA | — | 99.5% | 99.9% | 99.95% |
| Support | Community | Email + tickets | Dedicated + SLA | |
Overage
Usage beyond your plan is billed at the rates below. Prefer a hard cutoff? Set a spending cap per workspace in the console.
| Resource | Rate | Notes |
|---|---|---|
| Runs | $0.01 / run | Settled monthly |
| LLM tokens (input) | provider cost × 1.15 | Varies by model tier; shown live in the console |
| LLM tokens (output) | provider cost × 1.15 | Same as above |
| Storage | $0.10 / GB / month | Billed on monthly peak |
FAQ
What happens when I exceed my quota?
We don't hard-stop your traffic. Usage beyond your plan is billed at cost-plus overage rates, and you can set a hard spending cap per workspace in the console if you prefer a cutoff.
How are LLM tokens metered?
Every LLM node records a usage event on completion — input/output tokens, model, and cost. You can aggregate by workspace, time range, provider, and model; every cent is attributable to a specific node and run.
Do you support bringing my own keys (BYOK)?
BYOK and custom model providers are available on the Enterprise plan. Other plans use the managed gateway (Anthropic Claude, OpenAI, and more) routed by fast / balanced / quality aliases.
Can I self-host?
Yes. Enterprise includes a production Helm chart for Kubernetes, with support for externally managed PostgreSQL and Redis. Your data never leaves your infrastructure.
Will pricing change during the beta?
Current pricing reflects our public beta and may be adjusted based on real usage data. Any change will be announced in advance and won't affect the current term of an active subscription.
* Public-beta pricing; may be adjusted before general availability with advance notice.
Your next production agent starts on a blank canvas
The Free plan includes 100 runs and 500K tokens per month — enough to ship a real workflow. No credit card required.