Perplexica is an open-source, AI-powered search engine designed as an alternative to Perplexity AI. It aims to provide users with a privacy-focused, deeply insightful web search experience, leveraging local or cloud-based LLMs and a metasearch engine for up-to-date results.

How It Works

Perplexica utilizes SearxNG as its metasearch backend to fetch current web results, which are then processed by AI models. It supports various LLMs via Ollama or API keys (OpenAI, Groq, Anthropic) to understand queries, refine searches, and generate summarized answers with cited sources. It offers "Focus Modes" for specialized searches (academic, YouTube, Wolfram Alpha, Reddit) and is developing a "Copilot Mode" to actively refine search queries by visiting top results directly.

Quick Start & Requirements

• Docker: docker compose up -d (after cloning and configuring config.toml).
• Non-Docker: Install SearXNG, clone repo, configure config.toml, npm i, npm run build, npm run start.
• Prerequisites: Docker, Node.js (for non-Docker), API keys for cloud LLMs (optional), Ollama (for local LLMs).
• Setup: Docker recommended for simplified dependency management.
• Access: http://localhost:3000
• Docs: https://github.com/ItzCrazyKns/Perplexica

Highlighted Details

• Supports local LLMs (Llama3, Mixtral) via Ollama.
• Integrates multiple LLM providers (OpenAI, Groq, Anthropic).
• Features specialized "Focus Modes" for academic, YouTube, Wolfram Alpha, and Reddit searches.
• Uses SearxNG for real-time, up-to-date search results without manual indexing.
• Offers an API for integration into other applications.

Maintenance & Community

• Active development with several features already implemented and more planned.
• Community support via Discord server.
• Open to contributions via GitHub Issues.

Licensing & Compatibility

• The README does not explicitly state a license. Compatibility for commercial use or closed-source linking is undetermined.

Limitations & Caveats

The project is still under active development, with "Copilot Mode" noted as "in development." The license is not specified, which may impact commercial adoption. Linux users may need to adjust Ollama service configurations for proper API access.