Feature Request: Improving UX at Context Limit – "Summarize & Continue"

[drazenRD8]

Hello DeepSeek Team,

I would like to share a highly frustrating user experience that keeps occurring, along with a suggested solution.

The Problem: When a conversation reaches the maximum context limit (token limit), the user is abruptly interrupted with a very "cold" message: "You have reached the limit, please start a new chat." This happens even when moving to a completely new chat window (as was the case in my 6th consecutive conversation), which strongly suggests a bug in caching or token calculation.

This kind of abrupt cutoff is a poor UX because the user loses all context, the time invested, and feels like the conversation was "cut off at the knees."

Proposed Solution: Instead of an abrupt termination, please implement a "Summarize & Continue" feature.

How it would work:

1. When the system detects the limit is approaching, instead of cutting off, it would offer a button: "Conversation is nearing its limit. Would you like to summarize and continue in a new window?"
2. Upon clicking, the AI would automatically extract the key points, context, and current task status (e.g., "User is on Fedora, installing a game via Lutris, waiting for download to finish").
3. This summary would be automatically transferred to a new, clean window, allowing the user to continue exactly where they left off without having to re-explain everything from scratch.

This would significantly improve user retention and reduce frustration, especially during long, technical conversations where context is crucial.

I hope you will take this suggestion into consideration. Thank you.