Pricing

Start free. Scale when ready.

The core engine is open source. Add platform features, multi-provider VLM orchestration, and enterprise security as you grow.

Open Source

Free

Self-hosted core engine. Community support. Full pipeline access.

  • ■Core ingest + analysis pipeline
  • ■Single VLM provider
  • ■Structured JSON output
  • ■Zod schema validation
  • ■Community GitHub support
  • ■Docker deployment
Get Started
Pro

Contact Us

Managed platform with multi-provider VLM, SSO, and priority support.

  • ■Everything in Open Source
  • ■Multi-provider VLM orchestration
  • ■Automatic failover chains
  • ■SSO (SAML / OIDC)
  • ■Role-based access control
  • ■Priority email support
  • ■Cloud or self-hosted deployment
  • ■App marketplace access
Request Access
Enterprise

Custom

Air-gapped deployment, custom models, dedicated support, SLA guarantee.

  • ■Everything in Pro
  • ■Air-gapped / on-prem deployment
  • ■Custom VLM model integration
  • ■Dedicated support engineer
  • ■99.9% uptime SLA
  • ■Audit logging + SIEM export
  • ■Custom data retention policies
  • ■SOC 2 compliance documentation
  • ■Unlimited video throughput
Contact Sales
FAQ

Common questions.

Can I start with Open Source and upgrade later?

Yes. The open-source core is the same engine that powers Pro and Enterprise. Upgrading adds multi-provider orchestration, SSO, and platform features on top.

How is pricing calculated for Pro?

Pro pricing is based on video hours analyzed per month and the number of concurrent streams. Contact us for a quote tailored to your volume.

Do you offer a free trial of Pro?

Yes. We provide a 30-day evaluation with full Pro features, including a guided onboarding session with our team.

Can Enterprise be deployed completely air-gapped?

Absolutely. Enterprise supports full air-gap deployment with local VLM inference. No external network calls required.

What VLM providers are supported?

Open Source supports one provider of your choice. Pro and Enterprise support automatic orchestration across Gemini, GPT-4V, Claude, and custom models.