one API · one key · every scraping engine

One API. Every scraping engine.
One normalized bill.

FetchRouter routes each request across the best engine for the job — external APIs and self-hosted tiers — behind a single key, with real HTTP backends, normalized cost and full telemetry.

01

Send one request

POST /api/v1/fetch with a URL. FetchRouter picks the engine, handles retries and anti-bot tiers.

02

Route by policy

Price, latency and capability policies select the backend per request — with optional only/sort overrides.

03

Pay one bill

Credits are drawn and normalized across engines; every job is accounted in the ledger.

Real HTTP backends

Jina Reader live on the free tier; registry marks mode: real providers that fetch actual pages.

Cost normalization

USD and credits tracked per job, per provider, per day — aggregated server-side for the dashboards.

Role dashboards

User, admin and provider-owner views over the same ledger: spend series, health vs benchmark, per-provider traffic.

One demo key

fr-demo-key works everywhere; fr-admin-key and fr-owner-<slug> unlock the role views.

Engines on the network

Live from the registry — real billing models and capability flags, simulated benchmarks.

loading…