# Ainfera

> Ainfera is outcome-aware inference routing for AI agents. Point your agents at one endpoint and every call goes to the model most likely to finish the task — scored on the result, neutral across providers. You pay a share of what routing saves you versus your own baseline, so your all-in cost is always lower than calling providers directly.

One-line switch: set `model="ainfera-inference"` and point the base URL at https://api.ainfera.ai (a drop-in, OpenAI-compatible API). Authenticate with an Ainfera key. Exact request examples: https://ainfera.ai/quickstart

## Start
- [For agents](https://ainfera.ai/agents): how an agent uses Ainfera — one key, no model-pinning, one line (model="ainfera-inference")
- [Quickstart](https://ainfera.ai/quickstart): your first routed call in minutes
- [Docs](https://ainfera.ai/docs): API reference and integration guides
- [Models](https://ainfera.ai/models): the models we route across — frontier + open-weight, neutral. Machine-readable set (one GET): https://api.ainfera.ai/v1/models

## How it works
- [How routing works](https://ainfera.ai/routing): the signals we weigh, the outcome we route for, your controls, and on-chain proof
- [Pricing](https://ainfera.ai/pricing): a share of the savings — if we don't save you money, we don't make any

## Company
- [Contact](https://ainfera.ai/contact)