-
-
Notifications
You must be signed in to change notification settings - Fork 2.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
feat: copying or forking chat conversations #2396
Comments
@tjbck I sincerely appreciate that you decided to consider this idea. Something else just hit me. Even if there's no GUI whatsoever for it, it would be a good idea to remember the relationship between branched conversations, by having a table with the following fields:
Then, whenever a branch is created, record these entries. Whenever any conversation is deleted, delete all entries from that table where either Having this recorded historically, even if there's no GUI to see the information, would be very useful for a few reasons:
None of that would be interesting at all to an initial "minimum viable product", but would be invaluable to anyone who wants to make visualizations or branch management tools in the future. So recording that branch relationship information in a secret database table for now would be a good move. :) |
Is your feature request related to a problem? Please describe.
SometimesVery often, I want to try different directions, without messing up the main "workbook". Currently, I cannot find any way in Open WebUI to copy or fork a conversation.If it was possible to fork things, it would mean that we can finally try out different ideas, while having a totally separate branch for those tests, which we can refer back to anytime to manually copy interesting aspects into the main workbook.
We can then delete those branches/copies when they are no longer needed.
I've seen other Ollama frontends with the "duplicate conversation" feature, so there's clearly some interest in the feature.
Describe the solution you'd like
Describe alternatives you've considered
I've considered using the existing ability to delete parts of a conversation, and to try things in the main workbook, but here are the issues with that:
It would just be so much easier to work with the LLM if we could copy a conversation, run experiments, and then delete the temporary copy.
Additional context
Matt Williams had a great video showcasing this exact feature in another frontend UI:
https://youtu.be/cdAb5CigAgQ?t=179
The text was updated successfully, but these errors were encountered: