Skip to content

Perplexica

An open-source AI search engine that cites its answers

Visit official site

What Perplexica is

The platform is described as an open-source AI-driven search engine that searches the web and understands the question put to it. It uses machine learning methods such as similarity search and embeddings to refine results, and it works with local LLMs such as Llama3 and Mixtral.

What you can do with it

  • Get clear answers that cite their sources
  • Choose between Copilot mode in development and a normal mode
  • Pick from six focus modes such as academic search and writing assistant
  • Search images and videos while SearxNG supplies current information

Who it is for

  • Researchers and students use academic search mode to find articles and papers
  • Content creators use writing assistant mode to improve writing work
  • Technical developers use local LLMs for development and testing

Getting started

  • Make sure Docker is installed and running on the system
  • Clone the Perplexica repository from GitHub
  • Move into the directory that holds the project files
  • Rename sample.config.toml to config.toml

Pros & cons

✓ What we like

  • Image and video search are both supported
  • Six focus modes are described, including YouTube and Wolfram Alpha search

! What to watch out for

  • Copilot mode is described as still in development
  • The documented setup needs Docker running on the system

FAQ

What are the two main modes?

The material names Copilot mode, which is in development, and a normal mode.

Where do answers come from?

The engine uses SearxNG to fetch current information and returns answers with cited sources.

Last reviewed: 2026-11-15

More AI search engine tools

View all →

How we review