AI router config builder
Chain your own provider keys into ordered fallback lists, then export the chain as a config for opencode, LiteLLM, or any OpenAI-compatible client. Your keys and your traffic never touch our servers — this page only builds a list that stays in your browser until you export it.
Loading…
What it builds
A fallback chain is an ordered list of AI providers: the first hop is the primary, every next hop is tried when the one above fails, rate-limits, or goes down. Instead of editing YAML by hand, you pick providers from your own vault keys, drag them into order, and export a config your tools already understand.
Export formats
- LiteLLM YAML — a true cascading chain: one local endpoint, automatic failover across every hop.
- opencode.json — every hop wired as a provider so you can switch between them in the UI.
- OpenAI-compatible snippet — a shell setup for any client that speaks the OpenAI API shape.
- Raw JSON manifest — the route list itself, for anything else you want to wire.
- .env — the key material for the route, generated locally; configs only ever reference env vars.
No middleman, by design
Hosted AI routers proxy your traffic and meter you in credits. This builder never sees a request: it runs in your browser, stores routes in your own browser storage, and exports files that reference environment variables on your machine. You keep your own provider accounts and pay providers their listed price — many have real free tiers, so a chain can cost nothing at all. Manage keys in the key manager; browse all 42 providers; or read the OpenRouter alternatives comparison.
How it works
- Add provider keys in the key manager (or inline, below).
- Create a route and stack fallback hops in priority order.
- Pick an export format and generate — copy or download the config.
- Drop the config into opencode, LiteLLM, or your own client and run it.