Every chat runs on one model, and where the model offers reasoning levels, the reasoning effort sets how much it thinks before it answers. The model picker in the message box opens the model menu, where you choose both. This page covers the menu, what a choice saves for new chats, and what happens when a model is no longer offered.
Open the model list
The model control sits in the message box's control row. It shows the model's mark, its name and its effort level; when the row runs out of space it shortens to the mark alone, and the full choice stays in its tooltip and in the open list.
The /model command opens the same list for the chat. The two always agree: a change made in one is what the other shows next time it opens.
Choose a model
Click the model control in the message box, or type /model. The list opens on the model in use.
Pick a model. The list closes, and from the next request Oda uses the model you picked.
A switch applies from the next request. You can switch at any time, including while Oda is working: a request that is already running keeps the model and effort it started with. This differs from a chat with a Mind in DataMind OS, which keeps the model it started with (Models in chat).
The message box shows the model Oda will use next.
Picking the model that is already set just closes the list; nothing is sent.
While a choice is applied, the pending row shows a small spinner and reads Switching model…, and the other rows ignore clicks until the switch lands.
Set the reasoning effort
A model that reasons offers its levels in the Effort control at the foot of the list.
Open the list and highlight the model you want; the Effort control follows the highlighted model.
Choose a level. It applies at once, and the list stays open.
The levels are Off, Minimal, Low, Medium, High, Extra high and Max, as far as the model supports them. Default appears where a model carries no default of its own.
A model that does not reason shows no Effort control.
Changing the level for a model the chat is not using switches the chat to that model.
A request that is already running keeps the effort it started with.
From the keyboard
Up/Down, Home/End and PageUp/PageDown move through the rows; PageUp and PageDown jump between groups.
Type any letter or number to search the list.
Enter picks the highlighted model and closes the list. Tab moves between the rows and the Effort control, where Left/Right change the level.
Escape, or Shift+Tab from the rows, closes the list back to the model control. A click outside closes it as well.
Search and read the list
Models are grouped by maker. DataMind's own models come first; the other groups follow in the order the list gives them, not alphabetically. The group headings stay in view as you scroll, and with models from a single maker no headings are shown.
Above six models, a Search models 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, by design.
Each row shows the model's description, and a New badge on a recent addition.
The list comes from your company's DataMind OS.Settings > Models decides what appears in pickers: add providers, refresh their model lists, or hide models you do not want to see; see Models and providers and Models in the chat picker.
What Oda saves for new chats
An accepted switch is saved as the default. New chats start on that model. Until you switch one, new chats start on your company's default model, Kartlos Standard. If Oda refuses a switch, nothing is saved and the chat keeps the model it had.
The effort is saved together with the model. New chats that start on that model begin at the level you chose. If you only change the effort of a model that a preset started this chat on, the saved default stays as it was.
A model recorded when a chat is created is not saved. For example, an app that starts a chat for you can name its model: it applies to that chat's first request only and does not change the saved default.
A preset can name a default model of its own. A chat follows its preset's default model until you choose a model yourself or it runs its first request; after that, a preset change does not replace the model. See Create and edit agent presets.
When the list will not load
While Oda reads the list, the control reads Loading models…. When it reads the list again, the last list stays visible and the menu notes Refreshing model list….
If one maker's models cannot be read, its group shows {name} failed to load: {message} with Retry; the groups that loaded stay usable. In the /model list, a failed group reads Catalog failed to load: {message}.
An empty list reads No models available., and a search with no match reads No model matches your search.
A switch that fails shows Model operation failed: {message} and leaves the list open, so you can pick again; your chat keeps the model it had.
If another running copy of Oda is using the chat, the change is refused and Oda asks you to quit the other running instances; The message box explains that message.
When a model is no longer offered
A model can leave the list: the list your company offers can change, and models can be hidden from the pickers. Hidden models never appear in the picker or in /model.
Hiding does not switch a model off: it still answers a request that names it.
If the chat's model is no longer offered, the message box shows it as unavailable and 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.
No. Both open the same list for the chat, so a change made in one is what the other shows next; it applies from the next request.
Does the model I pick become the default for new chats?
Yes. When Oda accepts a switch, it saves the choice as the default, so new chats start on that model. Until you switch one, new chats start on your company's default model, Kartlos Standard.
I changed the model while Oda was working. Which model answers?
The switch applies from the next request. The request that is already running keeps the model and effort it started with, and the new model takes over from there.
Can I try a model in one chat without changing the default?
Not by switching in the picker: an accepted switch is saved as the default. A model recorded when a chat is created, for example by an app that starts the chat for you, applies to that chat's first request only and does not change the saved default.
Which effort levels can I choose?
Off, Minimal, Low, Medium, High, Extra high and Max, as far as the model supports them; a model that offers no levels shows no Effort control. Default appears where a model carries no default of its own.
I changed the effort and the model changed too. Why?
The Effort control belongs to the model highlighted in the list. If you change the level for a model the chat is not using, Oda switches the chat to that model; the new level applies at once.
Is the reasoning effort saved with the model?
Yes. The effort of the model that new chats start on is saved together with the model. If you only change the effort of a model that a preset started this chat on, the saved default stays as it was.
How do I choose a model with the keyboard?
The list opens on the model in use: Up/Down, Home/End and PageUp/PageDown move through the rows, and typing searches. Enter picks the highlighted model and closes the list; Tab moves to the Effort control, where Left/Right change the level. Escape, or Shift+Tab from the rows, closes the list.
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.
Part of the model list failed to load. What can I do?
Choose Retry where the failure shows. When one maker's models cannot be read, its group shows {name} failed to load: {message} and the other groups stay usable.
Why does Oda ask me to select a model before I can send?
The chat's model is no longer offered: it was hidden from the pickers, or it has left the list your company offers. Choose another model to carry on; Updates and troubleshooting lists the message and what to do.
Can I change the model of a subagent's chat?
No. A subagent's chat does not offer the model picker or /model: a subagent takes the decision recorded for the chat that started it. Choose models for Subagents covers subagent models.