Auto Router

Auto Router (model ID: openrouter/auto) is a meta-model that doesn't run inference itself — instead, it analyzes each incoming request and routes it to the optimal free model available at that moment. The router intelligently filters for models that support the features your request needs, such as image understanding, tool calling, and structured outputs.

This makes it the simplest way to get started with CallMissed on the free tier: just set your model to "auto" and the router picks the best available free model for your specific prompt. If your request includes an image, it routes to a vision-capable model. If you need function calling, it picks a model that supports tools. If you just need text generation, it picks the fastest available option.

The router considers factors like prompt complexity, task type, and model capabilities when making its selection. Free-tier rate limits apply: approximately 20 requests per minute.

On CallMissed, the "auto" model ID is an alias for openrouter/auto. It's available on the free tier and charges a small flat rate per call (2.00/10.00 per 1M tokens) to cover infrastructure costs, even though the upstream model inference is free.