Hi, first of all thank you for your work on FlowDown!
I'd like to propose a feature: In conversations, it would be helpful to display which model returns each message. Sometimes, multiple models may be used in multi-turn dialogues, and it's useful to see exactly which model responded to each message.
This can aid in debugging, understanding model performance, and comparing outputs during mixed-model usage scenarios.
Thank you for considering this suggestion!