-
-
Notifications
You must be signed in to change notification settings - Fork 1.1k
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
Multiple chat histories, support gpt-3.5-turbo #2
Comments
Those are two great ideas. I'll merge the pull request in a couple of hours, great point about the speed of the model. I find myself using the full context window of GPT-4 frequently, it's so good at chewing large amounts of data! Any idea on how to get the conversation sidebar? UX should be doable / easy, but I wonder where to store the data (there's no central storages/db right now) |
I think the lack of an external DB or storage is one of the selling points of the project. It allows more privacy for the users' prompts that are, in this way, shared only with the OpenAI API. |
How if we used some local storage of the browser (or similar, like IndexedDB) to store the conversations, on a per-browser basis? |
Yes, was thinking localStorage would be an OK way to do this, you could probably fit a lot of chat history in there if you compressed it. You could also set an expiry on the messages so the oldest ones are deleted to free up space. Another idea, along the same lines, is to clear the context of old messages once the context window is full, i.e. you hit 8k tokens and start getting |
Closed the original issue (3.5-Turbo) thanks to @koganei. Moved other issues out |
Added support for Markdown formatting in the chatmessages
Would be nice to have the chat history sidebar like ChatGPT. Also would like to be able to use gpt-3.5-turbo occasionally, since it's faster.
The text was updated successfully, but these errors were encountered: