The cheapest path for every model you call.
Route each call to the cheapest provider that can serve it, and fall back automatically when one fails. One config across OpenRouter, Kunavo, fal & Runware — lowest price per token, every time. Built for the Vercel AI SDK.
Each request takes the cheapest healthy provider — when one goes offline, traffic reroutes mid-flight to the next cheapest. No dropped calls.
−30%
on Anthropic + Google models via Kunavo
11+
image models, cheapest provider per model
auto
failover the moment a provider errors
USD
real per-call cost tracking, built in