You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, switching models requires going to the model tab in the Settings modal or using the slash command /model <model-name>. Switching characters can only be done with the slash command /character <character-name>.
I am thinking that it could be faster and easier to manage if there was a "Quick Controls" modal (launched from a button in the bottom of the window) so it is faster to manage chat controls without having to go to the Settings modal.
It is difficult for me to explain exactly with words, so I used AI slop to show what I mean. The example below has:
A Model selector: A dropdown to quickly switch the active model.
A Character selector: A dropdown to switch the active character (with an option to reset to default).
Some quick toggles for high-frequency settings, specifically:
Enable Thinking/Reasoning.
Keep Reasoning in Context.
quick-control-video-example.mp4
How I have it now is that the model selector is interacting with the existing /api/models, while the character selector builds on the /api/characters endpoint to provide a visual selection list instead of relying on slash commands. But for both, when selecting the value from the dropdown box it automatically runs the slash command (giving both auditory and visual feedback in the chat).
Sorry for not opening a PR, but the thing I got working is AI slop, and I don't want to pollute this project.
Currently, switching models requires going to the model tab in the Settings modal or using the slash command
/model <model-name>. Switching characters can only be done with the slash command/character <character-name>.I am thinking that it could be faster and easier to manage if there was a "Quick Controls" modal (launched from a button in the bottom of the window) so it is faster to manage chat controls without having to go to the Settings modal.
It is difficult for me to explain exactly with words, so I used AI slop to show what I mean. The example below has:
quick-control-video-example.mp4
How I have it now is that the model selector is interacting with the existing
/api/models, while the character selector builds on the/api/charactersendpoint to provide a visual selection list instead of relying on slash commands. But for both, when selecting the value from the dropdown box it automatically runs the slash command (giving both auditory and visual feedback in the chat).Sorry for not opening a PR, but the thing I got working is AI slop, and I don't want to pollute this project.
Thanks for your time.