MiniSearch is a privacy-focused, minimalist web search engine with an integrated AI assistant, designed for users who want to search the web and get AI-powered insights directly from their browser. It leverages WebLLM, Wllama, and SearXNG to provide a seamless, cross-platform experience without data collection.

How It Works

MiniSearch runs AI models directly in the browser using WebLLM and Wllama, eliminating the need for server-side processing for core functionality. It integrates with SearXNG as a meta-search engine to aggregate results from various sources. This client-side approach enhances privacy and reduces server costs, while caching models efficiently for quick access.

Quick Start & Requirements

• Install/Run: docker run -p 7860:7860 ghcr.io/felladrin/minisearch:main
• Prerequisites: Docker.
• Setup: Minimal setup time, primarily involves running the Docker container.
• Demo: https://felladrin-minisearch.hf.space

Highlighted Details

• Runs AI models directly in the browser using WebLLM and Wllama.
• Integrates with SearXNG for meta-searching.
• Supports setting MiniSearch as a default browser search engine.
• Offers an "Internal OpenAI-Compatible API" feature for sharing API keys securely.

Maintenance & Community

• The project is hosted on GitHub: https://github.com/felladrin/MiniSearch
• Contributions are welcomed via pull requests.

Licensing & Compatibility

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

Limitations & Caveats

The project's license is not specified in the README, which may impact commercial adoption. The reliance on a single Docker image for HuggingFace Spaces compatibility is noted as a design constraint.