Pricing API
What shipped
- `GET /pricing/rpc` — current price per million requests for every combination of vm, chain ID, JSON-RPC method, and routing strategy (economy vs performance). At launch, 168,000+ routes across 1,450+ chains.
- `GET /pricing/ws` — current websocket notification price per million, per chain ID and subscription type (newHeads, logs, newPendingTransactions).
Both endpoints are public and rate-limited per IP (30 requests per 60 seconds). Empty results return an empty JSON array, so you can rely on the response shape in code.
Try it
curl "https://api.routeme.sh/pricing/rpc?chain_id=1&method=eth_blockNumber"[
{
"vm": "evm",
"chain_id": "1",
"method": "eth_blockNumber",
"routing_strategy": "economy",
"price_per_million": 3
}
]Filter the RPC endpoint with vm, chain_id, method, or routing_strategy, and the WS endpoint with chain_id or subscription_type.
Why prices are public
RouteMesh prices are set algorithmically — driven by which providers currently have quality capacity on each route — not by a price list humans maintain. A static pricing page is a snapshot that goes stale the moment it is published. A public API returns the actual signal.
That also means you can make decisions with the same data we route on. Before you commit an integration, query the exact route you will be calling — chain, method, and strategy — and see what it costs right now.
What you can build
- Cost checks in CI — fail a build when a route you depend on drifts above a budget.
- Cost dashboards — poll the endpoint and track price movement per chain over time.
- Agent tooling — plain JSON with no auth makes it trivial for an LLM agent to price a call before it makes one.
- Provider comparisons — model your per-request spend across chains and strategies, or against a single-provider setup.
Details
Full reference, including parameters and error codes: Pricing API reference. How prices are derived: Pricing model.