Pricing built for production teams
Straightforward per-invocation pricing. No seat fees. No hidden egress charges. Scales with your actual usage.
For individual engineers and small teams evaluating OrchVynt in staging or development.
- All 4 orchestration primitives
- 7-day telemetry retention
- Community Slack support
- 1 environment
- Single-user HITL approval queue
For production engineering teams with live workloads requiring SLA and multi-environment support.
- All Starter features
- 90-day telemetry retention
- Email + priority Slack (48h SLA)
- Up to 5 environments
- Multi-user HITL approval queue
- SSO (SAML / OIDC)
- Audit log export (JSON / CSV)
- Uptime SLA 99.9%
For large engineering organizations needing self-hosted deployment, custom retention, and dedicated support.
- All Production features
- Self-hosted deployment (VPC / on-prem)
- Custom telemetry retention
- Dedicated Slack + named CSM
- Custom SLA (up to 99.99%)
- Private cloud deployment option
- Compliance reporting package
- Custom HITL workflow integrations
Frequently asked questions
Each agent call routed through OrchVynt counts as one invocation, regardless of token length. A single user request may trigger multiple agent invocations if your workflow chains multiple calls — each hop through OrchVynt is counted separately.
Cloud-hosted: Yes by default for observability, encrypted at rest and in transit. You can disable content logging and retain only routing metadata if preferred. Self-hosted (Enterprise): No — your data stays entirely in your VPC. OrchVynt doesn't call home with content.
OrchVynt includes a passthrough mode. If the control plane becomes unavailable, your agents can be configured to fall back to direct-to-provider calls automatically. This means OrchVynt's availability doesn't add to your application's downtime risk. You lose orchestration features (routing policy, budget enforcement), but calls complete.
Yes — self-hosted deployment is available on the Enterprise tier. We provide a Docker image and a Helm chart for Kubernetes. The self-hosted version has full feature parity with the cloud-hosted version and doesn't call home with your traffic data.
On the Production tier, overage is billed at $0.000085 per invocation above the 5M monthly base. Overage is calculated at the end of each billing month. You can set a soft cap with alerts at a configurable threshold so your team knows before hitting overage.
Start with the free Starter tier today
50,000 routed invocations per month, all four orchestration primitives, no credit card required.