Lumen API pricing

Pricing

One small platform fee for the router + the underlying model spend at cost. The router pays for itself the moment one easy prompt skips a frontier model. Priced to undercut Portkey and Helicone at every tier.

Free

Evaluate the router on real traffic.

$0/mo
10,000 routed requests / mo
  • 1 API key
  • lumen-auto cost routing
  • 8-provider failover
  • Basic cost dashboard
  • Community support
Start free

Starter

For solo builders and side projects.

$9/mo
$90/yr (2 months free)
  • 150,000 routed requests / mo
  • Multiple API keys + per-key budget caps
  • Quality-floor controls
  • Cost-saved reporting
  • Email support
Choose Starter

Scale

High volume. 8× cheaper than Helicone Team ($799).

$99/mo
$990/yr (2 months free)
  • 15,000,000 routed requests / mo
  • Everything in Pro
  • Priority routing lane
  • Dedicated-model add-on available
  • SSO add-on + governance pack
  • Higher rate limits + budget ceilings
Choose Scale

Enterprise

SLA, private deploy, compliance.

Custom
Annual contract
  • Unlimited routed requests
  • SSO / SCIM + audit export
  • Private / VPC deploy + SLA credits
  • Custom provider allowlist
  • Dedicated solutions engineer
Contact sales

How model spend works: the platform fee above covers the router, dashboard, audit log and failover. The underlying model tokens are billed at provider cost (or run on your provider keys on Pro+). Because Lumen routes easy traffic to cheap models, most teams see net spend drop even after the platform fee — that's the whole product. Looking for a manual, model-by-model OpenAI-compatible gateway instead? See our sister product EAS LLM Gateway.

How we compare: Portkey Production is $49/mo and Helicone Pro/Team are $79/$799/mo — all observability-first and you still pay model tokens on top. Lumen is cost-routing-first: same failover and audit, a lower platform fee at every tier, and the router actively cuts your token bill instead of just watching it.