Models
Choose which models appear in your client. Set context windows, capabilities, and variants.
Location: Settings → Models
The Models tab shows every model Jean2 knows about. The first model with a default tag is used unless a preconfig overrides it. Set defaults, sync with the community list, or define custom models.
Sync models
Click Sync Models to pull in the latest community list. You can preview changes before applying them, or overwrite your local list entirely.
Add a model
Use Add Provider to add a supported provider that isn't configured yet, or Add Model to define a custom model. Useful for providers like OpenRouter that expose hundreds of models.
Edit a model
Click any model row to open the editor:
| Field | What it does |
|---|---|
| Model ID | The identifier sent to the provider's API |
| Display Name | How it appears in your client |
| Context Window | Maximum input + output tokens the model accepts |
| Max Output Tokens | Hard cap on response length |
| Label | Budget / Standard / Premium, for organizing your list |
| Capabilities | Toggle what the model supports: text, images, video, PDFs |
Variants
In the Variant section, define different thinking levels for models that support it (like OpenAI's reasoning models). These appear as High / Medium / Low effort options when selecting the model in chat.
What's next
- Chat Controls: how to switch models in a session
- Preconfigs: bundle a model into an agent personality