# DSPy

> Run DSPy programs deployed with the official MLflow serving interface.

App · 1 action. Web page: https://docs.lodol.com/docs/integrations/dspy

## Actions

### Run program

Run a DSPy program served with MLflow. Inputs must match the saved model signature; model calls may be billed.

**Inputs**

- Inputs (`inputs`, object, required): MLflow scoring request matching your saved DSPy model signature, for example {"messages":[{"role":"user","content":"What is 2 + 2?"}]}.

**Outputs**

- `data` (object): Service response.

**Example output**

```json
{
  "data": {}
}
```
