Show HN: I've built a locally running Perplexity clone (github.com)

Article summary

The author has built a locally running Perplexity clone called LLocalSearch, which uses Large Language Models (LLMs) to search the internet for answers to user questions. The system is designed to be privacy-respecting and can run on low-end hardware. LLocalSearch uses a chain of LLMs to find answers and allows users to see the progress of the agents and the final answer. The project is open-source and available on GitHub.

Main themes

• Local search engines
• Large Language Models
• Privacy-respecting AI
• Open-source projects
• Natural Language Processing

What commenters say

• The use of local search engines and LLMs can provide a more private and efficient way to search for information.
• Some commenters are interested in using the project for offline search and querying of pre-installed websites and documents.
• There are concerns about the potential for anti-bot measures, such as CAPTCHAs, to limit the effectiveness of the search engine.
• Others see the potential for LLMs to replace human content creators, while some argue that AI can be a useful tool for content creators to augment their work.
• The project's ability to use a subset of the web and prioritize trusted sources is seen as a desirable feature by some commenters.
• There is interest in learning more about how the project works and how to get started with building similar projects.
• Some commenters are concerned about the potential for the project to be used for extensive searching, which could lead to issues with search engines' terms of service.
• The use of AI to parse and store large amounts of unstructured information is seen as a challenging but potentially valuable aspect of the project.