Pricing
Buy credits and pay for the work agents do, or license the platform and run it yourself.
Team
CREDITS
For teams that want agents working today.
Usage-based
Buy credits · pay for what agents do
Run unlimited agents and workflows
Unlimited seats, ideal for team-wide AI
Hundreds of models available
Connect agents to anything
Work from Slack, Teams, Claude, and ChatGPT
Enterprise
ANNUAL LICENSE
For companies that run the fleet on their own infrastructure.
Custom
Annual contract · from 50 agents
Self-deploy on your Kubernetes or on-premises
Bring your own model keys and negotiated rates
SSO and OIDC included
Sub-orgs with pooled credits and per-team usage
Private AI gateway connectivity
SOC 2 Type II certified and GDPR compliant
White-glove onboarding and Tier 1 support
Pricing FAQ
If you can’t find what you’re looking for, get in touch.
Credits are the unit agents spend as they work: 1 credit per agent wake, 1 credit per tool call, and model tokens at each model’s published rate. One credit is worth one cent ($100 is 10,000 credits).
Any message or event that wakes an agent. Each wake is 1 credit for the whole turn, no matter how long it runs. Tool and API calls the agent makes are 1 credit each, and model usage is billed separately per token at the model’s published rate.
Pricing is based on the work agents do, not on how many people use them. Add as many builders and users as you want.
Each model has configured rates in credits per million input and output tokens, and that rate covers everything model-side, so there is no separate provider bill. Claude Sonnet 5 costs 375 credits per million input tokens and 1,875 per million output tokens; GPT 5.6 Luna costs 125 and 750. So the 10,000 credits you get for $100 work out to roughly 26.7M input or 5.3M output tokens on Sonnet 5, or 80M input or 13.3M output tokens on GPT 5.6 Luna. Those figures are examples: what you actually spend depends on the model you pick and the work your agents do.
There is no self-serve trial. Book a 30-minute demo and our team will set up a pilot with you.
Self-deployment on your own infrastructure (Kubernetes or on-premises), your own model keys at your negotiated rates, SSO and OIDC, sub-orgs with pooled credits and per-team usage attribution, and a private AI gateway. Annual term with white-glove onboarding.