Pricing

Pay for what you route. Save as you scale.

Usage-based plans with included credits. No per-seat tax. Switch plans anytime.

Starter
$39
/mo · billed annually

For AI-native teams shipping their first production workloads.

  • 1M routed tokens / mo included
  • Up to 3 connected providers
  • Dynamic routing & failover
  • Core guardrails (PII, moderation)
  • Community support
Start with Starter
Enterprise
Custom
Volume + committed-use pricing

For regulated, multi-region organizations with procurement-grade needs.

  • Custom token commitments & discounts
  • VPC / on-prem deployment
  • Data residency & BYO-cloud
  • SAML, SCIM, granular RBAC
  • SOC 2 / ISO 27001 / HIPAA pack
  • Dedicated TAM & 99.99% SLA
Contact sales

All plans include the OpenAI-compatible API, streaming, observability dashboards and OpenTelemetry export. Overage tokens billed at transparent provider-cost + 8%.

Compare

What's in each plan.

Capability Starter Scale Enterprise
Included tokens / mo1M15MCustom
Connected providers3UnlimitedUnlimited
Dynamic routing & failover
Full guardrails + audit lineageCore
Unified memory bus
Priority lanes & quotas
SSO / SAML / RBACSSOFull
VPC / on-prem deployment
SLA99.9%99.99%
SupportCommunity24/7Dedicated TAM
Pricing FAQ

Questions, answered.

By routed tokens — input and output across every provider — plus any auxiliary calls (e.g. guardrail classifier tokens). Each plan includes a monthly token allowance; overage is billed at transparent provider cost plus 8%.
Yes. On Scale and Enterprise you can BYO provider accounts and pay providers directly; OMNOXA billing then covers only the gateway control plane and included tokens.
Traffic keeps flowing — no hard cutoffs on paid plans. Overage is billed at the end of the cycle, and you can set hard budget alerts and auto-throttles to cap spend.
Yes — every paid plan starts with a 14-day trial including 250k routed tokens, no card required. The Starter tier is also free for the first 30 days.
No. OMNOXA is priced by usage, not by seats. Add as many developers, tenants and API keys as you need.

Not sure which plan?

Share your expected volume and providers — we'll model your cost under each tier.