ItzCrazyKns / Perplexica

Perplexica is an AI-powered search engine. It is an Open source alternative to Perplexity AI

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

I don’t know why the search results often have nothing to do with the provided keywords? But if you search the same keyword directly through searxng, the result is correct.

edisonzf2020 opened this issue · comments

Describe the bug
A clear and concise description of what the bug is.

To Reproduce
Steps to reproduce the behavior:

  1. Go to '...'
  2. Click on '....'
  3. Scroll down to '....'
  4. See error

Expected behavior
A clear and concise description of what you expected to happen.

Screenshots
If applicable, add screenshots to help explain your problem.

Additional context
Add any other context about the problem here.

It really depends on the model you're using and how you ask your questions. If you're using open-source models like Ollama or Groq, the results might not be as good. But with OpenAI, they're usually pretty great. I suggest you try asking your query as a question if you haven't already, and maybe think about using a better model. I've tried both Groq and OpenAI a lot, and the results are good. And since its not an issue I would close this but we'll keep on upgrading the quality of answers over time as we move forward.

Please allow me to try again. I am using the Groq model. What I want to describe is not that the final answer of the language model is of poor quality, but that I find that sometimes there is a problem with the quality of the source of information returned by the entered keywords? That is to say, the search information returned from the search engine has nothing to do with the entered keywords. I think this is unlikely to be caused by the model. Is there any problem in the process of passing it to the search engine? Because I give the keywords directly to searxng, the search engine that this project relies on, the search results returned are no problem.
PS: This situation usually occurs when using Chinese as the language for inputting keywords, but is not found when using English.

Please allow me to try again. I am using the Groq model. What I want to describe is not that the final answer of the language model is of poor quality, but that I find that sometimes there is a problem with the quality of the source of information returned by the entered keywords? That is to say, the search information returned from the search engine has nothing to do with the entered keywords. I think this is unlikely to be caused by the model. Is there any problem in the process of passing it to the search engine? Because I give the keywords directly to searxng, the search engine that this project relies on, the search results returned are no problem. PS: This situation usually occurs when using Chinese as the language for inputting keywords, but is not found when using English.

I get it, Groq hosts open-source models and they are still little bad at multiple languages. We process the search using LLMs then a few algorithms to get the best result. Currently Chinese would be hard for the language models and hence the results are bad. I would recommend you to use English as we slowly reach the level of closed source models with open-source the quality of search results would also increase.