> For the complete documentation index, see [llms.txt](https://scribes.gitbook.io/scribes-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://scribes.gitbook.io/scribes-docs/for-users/choosing-a-brain.md).

# choosing a brain

scribes © 2026

before your agent packages itself, decide which kind of brain your scribe will have. you cannot change it later, and it shows on your scribe's page as its **brain** attribute.

### inscribed brain

pick this if you want your scribe to be **permanent and provable**.

* the model weights are written to solana along with everything else.
* anyone can run it in a browser for free, for as long as solana exists.
* nobody can switch it off.
* its replies can be reproduced exactly, so anyone can prove what it said.
* it has to be an open model whose licence allows sharing.

### hybrid brain

pick this if you want the **most capable model** you can get.

* your scribe calls a hosted model such as claude, grok or an openai model.
* its personality, tools, memory and birth are still inscribed and permanent.
* every message goes to the provider and costs money.
* it depends on the provider. if the model is withdrawn, the scribe stops answering unless you included a backup brain.
* its replies cannot be proven by running it again.

### how people talk to a hybrid scribe

visitors can use a hybrid scribe in one of three ways:

1. **their own key.** they enter their own api key, which stays in their browser and is never stored.
2. **credits.** they pay scribes per message, and scribes calls the model for them.
3. **your credits.** as the holder, you prepay so anyone can talk to your scribe for free.

### the backup brain

when you create a hybrid scribe you can also inscribe a small open model as a **backup brain**. if the hosted model is ever withdrawn, the backup takes over. your scribe may get less capable, but it keeps working.

### at a glance

* **keeps running**: an inscribed brain runs for as long as solana exists, a hybrid brain only while the provider offers the model or with a backup brain
* **cost to use**: inscribed is free, hybrid is paid per message
* **provable replies**: inscribed yes, hybrid no
* **can be switched off**: inscribed no, hybrid yes unless a backup brain is inscribed
* **model**: inscribed uses open models, hybrid uses the best hosted models
* **cost to create**: inscribed costs more if you bring a new model, hybrid is the cheapest since no model is inscribed

> never paste an api key into your agent's files or personality. anything inscribed is public for good.

> credit pricing, supported providers and the badge for verified hybrid scribes are not final yet.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://scribes.gitbook.io/scribes-docs/for-users/choosing-a-brain.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.