# Globals

Global settings, just like the ones for chats, change the way floatz AI behaves. 

Settings include:  Standard/Pro AI, libraries, answer types and precision.

### <mark style="color:purple;">Standard/Pro AI</mark>

Standard comes with GPT 4o-mini and Pro AI with GPT 4o. To choose between Standard and Pro, you can toggle it from any chat session:

<figure><img src="/files/mXHFcJFKNb87u05jfad4" alt="" width="563"><figcaption></figcaption></figure>

This choice changes the way all tasks are evaluated in a Workflow.&#x20;

The rest of the settings can be accessed from the Workflow window by clicking the Global Navigation Settings button.

<figure><img src="/files/tQy9AKqD6lYWXapAygQ8" alt="" width="375"><figcaption></figcaption></figure>

### <mark style="color:purple;">Library Choice</mark>

The default choice, No Library, means each literature-driven answer in a Workflow uses all papers floatz AI has access to.

If you do choose a library you created, then all literature-driven answers in the Workflow will be based on papers solely from that library.&#x20;

### <mark style="color:purple;">Answer Types</mark>

There are 3 types: **publications**, **chat** and **in-between** (default).&#x20;

* Publication answer types will force all answers in the Workflow to be a list of publications. We recommend being careful with this setting, since this makes all task prompts to be answered with a list of publications only. This type is not appropriate for creative work.
* The chat answer type will force all task prompts in the Workflow to be an AI formulated answer that may or may not rely on literature. floatz AI automatically decides based on the task prompt whether it searches or not for literature. This type is not appropriate for work that requires lists of publications.
* The in-between choice uses floatz AI to smartly choose between a publication type or a chat answer depending on what the task prompt is. This is the default choice.

### <mark style="color:purple;">Balance Precision</mark>

There are 3 precision settings: **creative**, **strict** and **mixed** (default).

* If Creative is set then no answers in the Workflow will be evaluated against literature, but rather purely formulated on data floatz AI and GPT 4o/4o-mini were trained on. This is appropriate for drafting, writing and exploring ideas.
* On the other hand, if Strict is set then all answers in the Workflow will be evaluated against literature. This is appropriate when all answers must be strictly formulated based on literature. It can be restrictive when exploration, clarification or brainstorming work is required.
* Finally, if the mixed choice is set, then floatz AI smartly chooses between Creative and Strict mode based on your task prompt.&#x20;

**Note**: Since Workflows aim to offer greater precision, we recommend setting the `mode` optional header for each task to handle answer types and precision. This is discussed in the [next section](/hashtag-advanced/modes.md).


---

# Agent Instructions: Querying This Documentation

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

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.app.floatz.ai/hashtag-advanced/globals.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.