The model picker

Every chat runs on one model. Where the model offers reasoning levels, the chat also runs at one of them. The model picker in the message box opens the Model and reasoning level menu, where you choose both. This page covers working the menu: searching a long list, changing levels with the keyboard, and what happens when a model is no longer offered.

Choose a model

1. Open the model picker in the message box. It shows the model of the chat.
2. The Model and reasoning level menu opens on the model in use.
3. Choose a model from the list. The choice applies straight away.

• An accepted choice is saved as the default for new chats. A chat that already has a model of its own keeps it; Model selection and defaults covers what a choice saves and when a switch takes effect.
• A leading default row pins nothing. A list can open with a row for the default, shown as the model it currently means, for example Default · Kartlos Standard. Choosing it leaves the chat on its default rather than a named model.
• While a choice is applied, the list holds. The pending row shows a small spinner, and the other rows ignore clicks until the switch lands.
• The same list opens with /model. Type /model in the message box to change the model of this chat; the commands are listed on Chat with Oda.

Set the reasoning level

A model that reasons offers its levels in the same menu, under Reasoning level.

1. Open the menu and press Tab. Focus moves from the model to the reasoning level in use.
2. Use the left and right arrow keys to change the level; each change applies straight away.
3. Press Enter to close the menu, or Shift+Tab to close it from the list.

• Only the levels the model declares are listed. The group leads with Default, then the model's declared levels, for example Off, High and Max.
• A level choice is saved together with the model. New chats that start on that model begin at the level you chose.
• A model that does not reason shows no level group.

Search and read the list

• Models are grouped by maker. DataMind's own models come first; the other groups follow in the order they arrive, not alphabetically. The group headings stay in view as you scroll.
• Above six models, a search box appears. Type to narrow the list. If the list already has focus, the first key you type goes into the search.
• Rows can show a context window, for example 200K or 1M: the amount of text the model can read at once. DataMind's own models do not show one; that is by design.
• A row can carry a small label, for example New for a recently added model.
• To add providers or hide models from pickers, open Settings > Models; Models and providers covers the page.

When a model is no longer offered

• If the chat's model is no longer offered, the message box stays blocked until you choose another model.
• The model picker stays active: open it and choose a model that is still offered to carry on.
• The message itself and what to do are listed on Updates and troubleshooting; Model selection and defaults explains what a change to the list means for your chats.

Questions, answered

How do I choose a model with the keyboard?

Open the model menu and use the keyboard throughout. The menu opens on the model in use: the up and down arrows move through the list, Home and End jump to its ends, and Page Up and Page Down jump to the next group; Enter chooses the model. Tab moves focus to the reasoning level, where the left and right arrows change the level straight away; Enter or Shift+Tab closes the menu.

The list is long. How do I find a model?

Once the list holds more than six models, a search box appears above it. Type to narrow the list; choosing a result works the same way. If the list already has focus, just start typing: the first key goes into the search.

What are the group headings in the model list?

Models are grouped under their maker's name, and DataMind's own models come first. The other groups follow in the order they arrive, not alphabetically. When the list holds models from a single maker, the headings are not shown.

Why do some models show a size like 200K?

That is the model's context window: the amount of text it can read at once. Some rows show one, for example 200K or 1M; DataMind's own models do not, by design.

What does the leading row for the default mean?

A list can open with a row that stands for the default, shown as the model it currently means, for example Default · Kartlos Standard. Choosing it does not pin a model: the chat keeps the default that applies to it.

I picked a model and a row is spinning. What is happening?

The picker is applying your choice. While it lands, the pending row shows a small spinner and the other rows ignore clicks; it clears by itself.

Do the model picker and /model show the same list?

Yes. Both open the same list of models, and a choice made in either one changes the model of this chat. The commands are listed on Chat with Oda.

Where does the list of models come from?

It lists the models your company's DataMind OS offers, each under the name it carries in that list. Settings > Models decides what appears in pickers: add providers, refresh their model lists, or hide models you do not want to see. Models and providers covers the page.