Generation is slow or times out: Context generation depends on how many source conversations are being processed. If you've selected many sources, generation may take longer. Try selecting fewer, more relevant sources.
Output quality: Generation is only as good as the conversations it has to work with. For best results:
- Select specific, relevant transcripts as sources using the @ picker
- Be specific in your request ("generate a brief on our pricing discussions from the last two weeks" vs. "summarize everything")
Generation stops unexpectedly: Press Cmd+N to start a new chat and try again. If the issue persists, contact us.