Typing Mind

I often want to ask clarification or dive deeper on part of a response. I would like to be able to create a quick side/sub-thread - without losing the main thread. I for sure don't want to send the whole context (which is often 100K+) just the response I'm asking a specific follow-up on.

Can we have conversations where we can choose multiple bots to talk to - similar to Silly Tavern? https://sillytavernai.com/ For example, then we could have brainstorming sessions with different bots with different skills, or learn languages in a room with multiple bots for more immersion. This will also open up the program to more people.

When the sidebar is collapsed, the chat name is not visible. Therefore, having the chat name (and possible folder name) at the top of the chat is beneficial. Or implement a breadcrumb.

Allow user to edit name, tag, add to folder in one place. Example Pain: user searchs for chats with text "ABC" - renames it...but once renamed, it disappears. Then have to search to then take 2nd action to put into a folder

Add multiple sorting options to the left panel interface to facilitate conversation history management. These would include sorting by creation date, modification date, or last accessed date. At present, selecting an old chat moves it to the top of the history, complicating its reading and categorization. The new sorting features would enable a more efficient and organized user experience.

Please let us bookmark a comment/reply in the chat and access all bookmarked comments in a separate area. Example: You iterated to and fro with the AI and finally made a roadmap that you're happy with, you bookmark that roadmap. Example: AI generated a code snippet that you know you'll need to access repeatedly, you bookmark that. Example: Something you don't want to act on atm but want to visit back later, you bookmark it.

Like the assistant replies - I want to have proper code section in user blocks

As I find myself using these agents more and more every day for simple, one-off tasks and questions, my conversation history gets messy. Could there be a mode, like a dedicated conversation, (optionally default) that isn't saved and prompts and responses more than X old self-destruct?

I use GPT-4-32k through OpenRouter a lot, and this is a slight annoyance, especially with how expensive the token usage adds up for that model. I know it would be hard to get the exact cost of each prompt, but an estimation would be great. Would it be possible to include tokens utilized next to a prompt/reply and the estimated cost?

It would be awesome to add ability to select a different model for the title and keyword generation instead of using the model of the conversation. For example, when using opus, it would be cheaper and faster to use haiku to generate the chat title and keywords.