Feature suggestions

It is not possible to modify the text output by Deepseek

There should be a setting where "Deeply Thought" dialogue is automatically collapsed when it is finished. While I understand the benefit of having the reasoning incorporated into the chat, I don't have any desire to see it or read it.

The app's theme color lacks a sense of depth; the light theme is just a vast expanse of white, while the dark theme is simply a whole swath of gray-black.

When chatting, I like to generate multiple conversations per conversation. I want to keep branch records, but I always mistakenly touch the “New Topic” button at the bottom, causing all side records to disappear. I'd like to be able to close the “New Topic” button freely. Or open a new topic to keep previous regeneration records

Only deep thought, no answers Deep thinking is not shown; the answer lies directly in deep thinking

Please ignore grammar mistakes because I am not an English speaker,thank you. Please add user-defined params in the http body,http head and http query,so that users can channel some settings in some special models. For example,"max tokens" is not availble for deepseek models. Having added function,we can realize the need by putting a "max_token" in the http body.

Add a toggle option in Settings allowing users to choose AI response generation methods: ​Character-by-character Streaming (current mode): AI responses appear in chat word by word, simulating real-time typing. ​Full-message Display (proposed mode): AI completes generation in the background before instantly displaying the entire response in the chat window.

I hope it will update automatically

Currently, most large models can set the output max_tokens, so that the maximum amount of content output by the large model is controlled by the user. I'd like this feature to be added to the advanced settings, at the same level as contextual message limits, rigor and imagination (temperature).