> For the complete documentation index, see [llms.txt](https://docs.protoboard.xyz/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.protoboard.xyz/protobot/models-and-reasoning.md).

# Models and reasoning effort

Choose a ProtoBot model and reasoning effort, and understand which models need a Pro plan and how they affect Energy use

## Goal

Choose the model and reasoning effort that give you good answers without using more Energy than you need.

## Choose a model

Click the model button under the message box. The list shows each available model with its price per million tokens, and badges:

* **Default**: the model new sessions use.
* **Pro**: available on the Pro plan. On Free, choosing it opens the upgrade options.

The model list changes as new models are released, so the menu is the place to check what's available. You can search it with **Search models...**.

<figure><picture><source srcset="https://3535233690-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHc6ZGiGpKHrs578383AS%2Fuploads%2Fgit-blob-fc2b74ac1bf4693d5b86d060bee1ba7c8853426d%2Fprotobot-model-selector-dark.png?alt=media" media="(prefers-color-scheme: dark)"><img src="https://3535233690-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHc6ZGiGpKHrs578383AS%2Fuploads%2Fgit-blob-79f5bdcd439a47571cbbddd8bb22453cf92038c2%2Fprotobot-model-selector-light.png?alt=media" alt="The model menu listing several models with Default and Pro badges and their input and output prices"></picture><figcaption><p>The model menu.</p></figcaption></figure>

## Choose a reasoning effort

The **Reasoning effort** button sets how long the model thinks before it acts:

| Level      | Use it for                                               |
| ---------- | -------------------------------------------------------- |
| **Off**    | "No extended thinking", fastest replies. Quick questions |
| **Low**    | Routine edits                                            |
| **Medium** | "Balanced reasoning (recommended)". The default          |
| **High**   | Hard design problems. Slower and uses more Energy        |

<figure><picture><source srcset="https://3535233690-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHc6ZGiGpKHrs578383AS%2Fuploads%2Fgit-blob-3ae8941958be26da7723f2d5624c2ecc96ca06bc%2Fprotobot-reasoning-dark.png?alt=media" media="(prefers-color-scheme: dark)"><img src="https://3535233690-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FHc6ZGiGpKHrs578383AS%2Fuploads%2Fgit-blob-e2031564dc6d02f65255cdae71c3ff26c6a3844c%2Fprotobot-reasoning-light.png?alt=media" alt="The Reasoning effort menu with Off, Low, Medium and High"></picture><figcaption><p>Reasoning effort.</p></figcaption></figure>

## How they affect Energy

More capable models and higher effort cost more per message. A good habit is to plan with a strong model, then switch to the default model and lower effort for routine follow-ups. See [Energy and usage](https://docs.protoboard.xyz/protobot/usage-and-energy.md).

## Related pages

* [Modes](https://docs.protoboard.xyz/protobot/modes.md)
* [Plans and billing](https://docs.protoboard.xyz/account/plans-and-billing.md)
* [Energy and usage](https://docs.protoboard.xyz/protobot/usage-and-energy.md)
