> ## Documentation Index
> Fetch the complete documentation index at: https://docs.clickml.app/llms.txt
> Use this file to discover all available pages before exploring further.

# Model Selector component

> Choose a specific saved version of a trained model from your ClickML Studio workspace to load into a pipeline for inference or evaluation.

Every time you run a model component, ClickML saves a new **version** of that model. The **Model Selector** lets you pick which version to use downstream — useful for comparing model runs or rolling back to a previous configuration.

## Configuration

| Option      | Description                                                                                                    |
| ----------- | -------------------------------------------------------------------------------------------------------------- |
| **Version** | The model version number to load. Versions are created automatically on each run and numbered starting from 1. |

## Input / Output

|        | Type                                 |
| ------ | ------------------------------------ |
| Input  | Trained Model                        |
| Output | Trained Model (the selected version) |

## Notes

* If no version is specified, the most recently trained model is used.
* Model Selector is optional. You only need it when you want to explicitly target a historical run.
* Place it between a model component and the Inference component.

```
Random Forest → Model Selector → Inference → Evaluation
```