Hello OpenAI team and community,
I’d like to propose a feature enhancement that would be a game-changer for power users: Permanent and Unlimited Chat Pinning in the ChatGPT conversation history panel.
Feature Summary:
Pin any conversation to the top of the sidebar for quick access
No limit on the number of pinned chats
Pins remain permanent until manually unpinned**
Optional enhancements:
Drag-and-drop reordering
Color tags or labels
Filter/search within pinned chats
This functionality would vastly improve workflow for users managing multiple ongoing threads, projects, or long-term research discussions.
While that doesn’t happen, here is a small trick:
- Create a project, and call it “pinned chats” or something alike
- Click on the dots (…) of your chat, and send them to your “pinned chats” project, and they will all be there in a single place for your convenience!
2 Likes
In facts, I got an infinite chat with GPT.
Maybe i’d just… Spammed it way to many prompts since i lauched that chat 
(~100 to 400 a day, since almost a week and a half)
And it keeps its narrative coherence.
It just deletes all messages that were sent/replied from the day its memory saturated, by doesn’t matter because I only use to reinforce contextual memory on my GptPlus account.
And @aprendendo.next is right, Projects (folders) are a good way to pin any chat on top of your history, or you just have to clearly label your “infinite” chat, to be able to easily search it later.
GPT got many features that aren’t explicitely shown on its app.
For exemple, you can easily use it to transfer files (in nearly any format, jpeg, pdf, apps…) from a device to any another.
Simply have to connect those 2 devices to the same GPT account.
Here is the way : launch a chat that you forbid to reply anything than “Received” or “ok”
Will avoid the chat to become boring with unncessary analysis of the file.
You sent you file from the first device, connect to the second, and finally download the file toute the receiving device.
It’s only a matter of little tricks, and dont know if OpenAi could really document all of this little tricks on an FAQ, cause each one is strongly bond with the user’s doings on his account