Pricing

Pay for outcomes, not seats.

Plans scale with active agents. We don't charge per user — your whole team can use any tier. Cancel any time during your trial.

Starter
Custom

For solo founders testing the waters. Get to "AI is doing real work" inside week one.

Talk to us
  • 3 active agents
  • Cost Efficient model tier
  • Slack + email integrations
  • Community support
Most popular
Growth
Custom

For small teams running multiple departments. Most operators land here.

Talk to us
  • 20 active agents
  • Smart + Power model tiers
  • 400+ integrations via MCP — on the roadmap
  • HITL approval queues
  • Priority support
  • Custom REST endpoints
Enterprise
Custom

Decentralised clusters, your VPC, your terms. Regulated-industry posture included.

Talk to us
  • Unlimited agents
  • Self-hosted gateway / your VPC (roadmap)
  • Full audit log export + per-tenant isolation
  • Role-based access · SSO + HIPAA BAA (roadmap)
  • Dedicated success engineer
  • 99.95% SLA

All plans are custom-priced during private beta · talk to a founder, not a sales rep

Full comparison

Every plan, side by side.

Starter Growth Enterprise
Active agents320Unlimited
Model tiersCost EfficientCost · Smart · Power · FreeAll + custom
Integrations15 core400+ via MCP · on the roadmap400+ via MCP · custom REST
Browser agent—✓✓
HITL approval queuesBasic✓ Full✓ Policy library
Business Brain memory1 GB25 GBUnlimited
Audit log30 days1 yearIndefinite + export
PII redaction✓✓✓ + custom patterns
SSO (SAML / OIDC)—RoadmapRoadmap
Self-hosted runtime——Roadmap
HIPAA BAA—RoadmapRoadmap
SupportCommunityPriority chat + emailDedicated SE · phone · SLA
OnboardingSelf-serveFounder-led (1h)White-glove (2 weeks)
Pricing FAQ

Why no checkout page.

We talk to every customer before they pay. AI workforces aren't a self-serve purchase yet — the setup matters too much to get wrong.

What counts as an "active agent"?
An agent that ran at least one task in the last 30 days. You can have unlimited dormant agents in your roster — only the working ones count. Pause an agent and it stops counting immediately.
Are there usage fees on top?
Most plans include enough model usage for typical workloads. Heavy users get a clear monthly usage report and can either tune model routing (the platform suggests this automatically) or bump up a tier. No surprise bills — usage that would push you over the cap requires explicit approval.
Can I bring my own model API keys?
Yes, on Growth and Enterprise. Plug your own OpenAI / Anthropic / Azure / Bedrock keys and we pass usage through at-cost. You get the orchestration; the model bill goes to you directly.
What's the difference between Growth and Enterprise?
Enterprise adds three things: self-hosting (your VPC or on-prem), advanced security posture (full audit-log export, per-tenant isolation, and custom security policies), and a dedicated success engineer. If your security team is going to be involved, you're probably an Enterprise customer.
How do I get started?
Access is invite-only during private beta. Request access and a founder helps you set up your first agents inside the first hour — there's no self-serve checkout yet. Pricing is custom and shaped around your fleet, and your data exports cleanly if you decide to walk away.
Can I downgrade?
Yes, any time, prorated. Pause agents you don't need and the system suggests when you've drifted below your tier.
Do you discount annual or non-profits?
Annual is 17% off (two months free). Registered non-profits get 30% off any plan — proof of status emailed to [email protected].

Talk to us. Pick a plan after.

A 30-minute call with a founder. We'll match you to a tier and show you the path to your first three agents.

Get started →