It just my thought. Since, Deepseek is opensource. Can Fedora ever use it? Local chatbot like Windows 11. Runs on local pc?
Deepseek is just one model among many.
I would be interested in something like ollama being able to run via Fedora repositories. I would not support Fedora packaging any particular model or the other.
Personally, I think LLMs are mainly elec-trickery. Their use cases are limited to constrained data-sets and far overblown at this point.
AI products from Chinese company DeepSeek is also majorly censoring topics with any controversial connection to its home government. DeepSeek is avoiding answering questions about topics commonly censored by the Chinese government, including human rights violations, government critiques and more.
You can use DeepSeek or any model available on Ollama through the Flatpak app Alpaca, a GTK4-based GUI for Ollama. The app functions as you would expect from a chatbot on Windows 11 with minimal configuration, assuming you know how to install the NVIDIA driver and Flatpak already.
The only issue I encountered was KDE desktop crashing, so I switched to GNOME.
Avoid using DeepSeek v2, which is included by default in Alpaca’s model manager, because it is much more heavily censored. Instead, manually add the deepseek-r1 model by name.
Have you tested yourself if not test it.
What this censorship is talked about is based on
Deepseek.com
If you run the models locally on your system it does not censor anything as far as i have tested.