AI Configuration
Model Configuration
Enable models and set workspace defaults in AINexLayer
Model configuration controls what appears in the chat composer and which models are used for specific roles.
Open Models settings
Workspace settings → Models (or Manage models from the composer).
Enable models
- Confirm a provider connection exists and is healthy.
- Enable the models you want available in chat.
- Return to chat and pick a model from the selector.
Disabled models do not appear as selectable options.
Auto vs specific model
- Auto — lets the product pick an appropriate enabled model for the request.
- Specific model — pins that chat to one enabled model.
Switch anytime from the composer; it does not require leaving the conversation.
Role defaults
In Models settings you can assign defaults for capability roles such as:
- Chat — main conversational model
- Vision — when image understanding is needed
- Image generation — when image creation is available
Roles appear when matching enabled models are available.
Tips
- Start with one solid chat model enabled, then add others as needed.
- If chat errors mention the model or provider, reopen Models and verify the connection.
- Workspace model choices are per workspace — switch workspaces if you need different stacks.
