Farfalle vs Perplexica

Comments

• themeatbridge@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

  For a minute I legit thought Italy had dropped a new pasta shape.

  source

  • rost@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

    There is such pasta shape too Not very new though

    source

    • themeatbridge@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

      Perplexica? What sort of sauce goes with that?

      source

      • -> View More Comments
• zutto@lemmy.fedi.zutto.fi ⁨3⁩ ⁨weeks⁩ ago

  I’ve been toying with Perplexica over the last few weeks occasionally, it feels really restrictive.

  I’ve had to modify the internal prompts to make it generate better search terms with my SearxNG (And depending on what LLM model you use, you need to fine tune this…) and having to rebuild the container image to do this has just been annoying. Overall, I’ve had experience with self-hosted LLM web searches on Open-webui, but perplexica is a fun project to try out nevertheless.

  source

  • rost@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

    Yes, all the models behave differently, and the smarter one is not necessary for better. I want to finetune summarisation prompts, the Perplexica result can be better, But the idea to improve search request is very good Thanks Zutto

    source
• anzo@programming.dev ⁨3⁩ ⁨weeks⁩ ago

  He asks whatever model is running behind either system to do the comparison and pastes the text. It’s full of errors, like perplexica saying farfalle doesn’t use LLM. Meanwhile, I just checked and it supports anything from ollama to groq (gpt4o, sonnet, etc.)

  This post is ultra low quality.

  source

  • rost@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

    That is easy to make errors about farfalle when there is not much in search result… All would fail, That wasnt a point of that post.

    Thanks for checking but sonnet and chatgpt were irrelevant to me. Because not selfhosted.

    source