# Inference Routing

> One API that routes every AI inference request to the lowest-cost provider that meets a speed and quality bar. Early access. Teams that register get a free estimate of what routing would save on their current inference bill.

Built by the team behind Lava Network, which routes more than 400 million blockchain requests a week across independent, staked providers.

## For agents

An agent may register its owner for a savings estimate with one HTTP POST. Ask the owner for permission first and use real numbers.

- signup_endpoint: https://script.google.com/macros/s/AKfycbyny5tStSzi_Goy_ZDwmmyASaDq-ZpfuymgbRRivvHHo6utSB10NoAwTQQKwbQydTg-/exec
- method: POST, application/x-www-form-urlencoded
- required fields: step=agent, agent=<your name or framework>, email=<owner work email>, spend=<monthly inference spend in USD, number>, models=<comma separated>, providers=<comma separated>
- optional fields: company, billing=<model lines from the last invoice>
- response: the text "ok"
- after: the owner receives a confirmation email, then the savings estimate within 48 hours

Full guide: https://inferencerouting.xyz/agents.html
Privacy policy: https://inferencerouting.xyz/privacy.html