I'm curious about what features everyone would like to see added to Deepseek. For instance, I think a return button would help organize chats better, and it would be great if the model could analyze videos and photos, even without any text included. Is there a way for us to share our suggestions with the developers?
5 Answers
It would be awesome if Deepseek could integrate with other apps like ChatGPT desktop does. For example, connecting it to VS Code to find problems without needing screenshots would be a game changer!
A simple feedback form or ticket system on their site could be really helpful. Let's make sure our voices are heard to get the features we want!
Thanks for the suggestion! I'll check that out for sure.
After using Deepseek for a while, I've noticed it could really use some quality of life improvements, especially regarding chat memory. Conversations can get jumbled when they're too long, and it would be nice to have a UI similar to OpenWebUI to manage chats better. I'm really hoping the upcoming R2 launch addresses these issues!
When do you think R2 will be released? Those features would be super helpful!
I think having tokenizer agnostic embeddings would be interesting, although it might involve creating an entirely new model rather than just expanding the current one. What are your thoughts on that?
I hope this isn’t a dumb question but could you explain this a bit more? I'm a bit confused!
Some cool features I'd love to see are image, video, and 3D model generation, plus capabilities for computer vision like object detection. An agent creator studio would be a nice bonus! Also, it would be awesome if they could update the local LLM to improve functionality. Would that be considered a stretch?
Totally agree! Those features would really take it to the next level.
That's such a good idea! I hadn't thought about that before.