Skip to main content

Pricing

Self-host the kernel. Pay for help and managed operations.

Stage 1 enforcement is free to self-host when tests pass. Optional Success Package helps with install; Stage 2–3 managed capabilities flip to current as they ship.

Available nowshipping with ActantOSComing soonon the product roadmap

Self-Host Open Core

Free

Everything needed to run the v1.2.0 Enforcement Kernel and Stage 2 ops on your own infrastructure, without artificial agent limits.

  • Policy engine (Cedar rules)Available now
  • Human-in-the-loop approvalsAvailable now
  • Postgres hash-chain auditAvailable now
  • Docker self-host deploymentAvailable now
  • Optional Slack + webhook approvalsAvailable now
  • Ops metrics, filters, policy dry-runAvailable now
  • Pi adapter and optional MCP gatewayAvailable now
How to use at work

Success Package

Contact us

Optional fixed-scope install and policy setup help for one real coding-agent workflow. Not a gate — the open-core kernel already ships when tests pass.

  • Kickoff and policy posture setupAvailable now
  • Limited office hours and async supportAvailable now
  • Install and first workflow greenAvailable now
  • Docs-aligned configurationAvailable now
  • No custom feature developmentAvailable now
Request help

Managed Platform

Custom

Commercial managed multi-tenant and Stage 3 isolation/evidence products. Self-host Stage 1–2 (Compose + optional OIDC ops auth) is free today — see /v1 and docs/HOSTED.md.

  • Managed multi-tenant platformComing soon
  • Enterprise SSO at managed scale (SAML, OIDC)Coming soon
  • gVisor / Firecracker isolationComing soon
  • WORM audit export (S3 Object Lock)Coming soon
  • SIEM connectors (Splunk, Datadog)Coming soon
  • Custom SLA & DPAComing soon
Talk to Sales

Success Package is optional help, not a ship gate. Managed-platform pricing follows Stage 2–3 capabilities as they are built and tested. Start with How to use ActantOS at work, or contact us for pricing and a technical assessment. Billing questions: billing@actantos.com.

Pricing FAQ

Common questions about ActantOS pricing and deployment.

It tests three bounded runtime-safety hypotheses: shared approval admission under concurrency, coupled approval–decision–effect safety under declared failures, and omission-aware effect evidence. These are research targets built on the v1.2.0 enforcement baseline, not claims that broad MAS novelty or exactly-once external effects are already proven.
ActantOS sits between your AI agents and your internal/external tools. It acts as an intelligent proxy that intercepts tool calls, evaluates them against your defined policies, and either blocks, pauses for approval, or allows the action.
As agents become more autonomous and capable of taking actions (like writing to databases or calling external APIs), the risk of hallucinations or malicious prompts causing harm increases. ActantOS ensures agents stay within strict boundaries.
No. ActantOS is completely model-agnostic. It works with OpenAI, Anthropic, Gemini, or any open-source model. It sits at the application/tool layer, not the generation layer.