Feature Request: Auto-continue for truncated outputs (Seamless Generation)

Dear team,

I would like to suggest an improvement for the interface. When requesting long responses, such as code blocks, the text often truncates due to the token limit, forcing the user to manually type "continue". This interrupts the workflow.

I have noticed that the lmarena platform recently solved this issue in their new interface. I am unaware of the technical details behind their implementation, but they now successfully generate long texts continuously without cuts.

Would it be possible to implement a similar solution on Venice? Automating the continuation of long responses would significantly improve the user experience.

Thank you for your work.