Models & Routing

routing.run exposes the models in its live catalog across multiple providers. Each model is identified by a unique ID and routed through its configured provider chain.

Model IDs are plain identifiers like deepseek-v4-pro. routing.run handles provider selection and uses another configured route when one is available and the request can be retried safely.

Browse the Model Catalog on the dashboard for the full list of available models with pricing and context limits.

When you send a request, routing.run selects a healthy route for the requested model. If another configured provider is available, a failed attempt can fail over before returning an error. Streaming requests can only switch routes before content has been sent to the client.

This happens transparently — your code does not need to handle provider selection or failover logic.

Call GET /v1/models to list every configured model and its public metadata. This catalog route does not require an API key:

curl https://api.routing.run/v1/models