Conversation

Notices

1. Embed this notice
   kaia (kaia@brotka.st)'s status on Saturday, 21-Sep-2024 01:30:45 JST kaia
   it's important to use LLMs a lot, so you instantly sniff out what is AI generated. this is a stark contrast to not using LLMs, a testament to people's reluctance to adapt.
   • Embed this notice
     Bricky (thatbrickster@shitposter.world)'s status on Saturday, 21-Sep-2024 01:59:54 JST Bricky

     in reply to
     @kaia Need me a self-hosted LLM to help me work. My executive dysfunction would appreciate a 'thought partner' when processing thoughts.
     kaia likes this.
   • Embed this notice
     Sally (sally@shitposter.world)'s status on Saturday, 21-Sep-2024 01:59:59 JST Sally

     in reply to

     • Bricky
     @thatbrickster
     @kaia
     
     
     https://github.com/nomic-ai/gpt4all
     kaia likes this.
   • Embed this notice
     kaia (kaia@brotka.st)'s status on Saturday, 21-Sep-2024 02:02:26 JST kaia

     in reply to

     • Bricky
     @thatbrickster lain and moon might be able to give more local recommendations, but I really like Mistral Nemo because it's quite uncensored. our frens use Llama, which I don't really like
     
     if you need to code and don't mind paying, Claude AI is very good. ChatGPT with its experimental model might be better now, I have not tried.
   • Embed this notice
     on-lain ✔ᵛᵉʳᶦᶠᶦᵉᵈ (lain@lain.com)'s status on Saturday, 21-Sep-2024 02:06:44 JST on-lain ✔ᵛᵉʳᶦᶠᶦᵉᵈ

     in reply to

     • Bricky
     @kaia @thatbrickster mistral-small is also good, the new qwen2.5 models too
     kaia and Sick Sun like this.