Suits self-hosters and developers who want AI search they control; you should be comfortable running a Docker container and entering model API keys or an Ollama URL.

What can you build with Vane?

Run a private alternative to Google or Bing that answers questions with cited sources. Ask questions about your own PDFs and text files alongside web results. Research a topic in Quality mode, limited to academic papers or chosen websites. Add a local AI search endpoint to your own app through Vane's search API. Pair it with a local Ollama model to run AI search without sending queries to a cloud LLM.

ItzCrazyKns /

Vane

Self-hosted AI answer engine that searches the web and cites its sources

37.7k stars
681 stars since 3 Oct
TypeScript
Updated 2 months ago
Some experience
Models & local AI

Open on GitHub (opens in a new tab)

Vane is a web app you run yourself that answers questions the way hosted AI search products do. You type a question, it searches the web, and a language model writes an answer with links to the sources it used. The repo's topics mention Perplexica, which appears to be the project's earlier name.

Under the hood it uses SearxNG, an open-source meta-search engine, to gather results from several search engines. The model then reads those results and writes the reply. You pick the model: a local one through Ollama or Lemonade, or a hosted one from OpenAI, Claude, Gemini or Groq. The Docker image bundles SearxNG, so there is little to set up.

The main difference from a hosted AI search product is control. Your history is saved locally, you choose the model, and you can add Vane to your browser's search bar or call it from your own code through an API.

What it does

Choice of model providers
Works with local models through Ollama, or with OpenAI, Anthropic Claude, Google Gemini, Groq and others. You can mix models.
Three search modes
Speed mode for quick answers, Balanced for everyday use, and Quality for deeper research.
Source and domain control
Search the web, discussions or academic papers, or limit a search to specific websites.
Files, images and widgets
Upload PDFs, text files and images and ask questions about them. Image and video search is included, and cards appear for things like weather, calculations and stock prices.
Search API and browser integration
A documented API lets other apps run searches. You can also add Vane as a custom search engine in your browser.

What you could build

Run a private alternative to Google or Bing that answers questions with cited sources.
Ask questions about your own PDFs and text files alongside web results.
Research a topic in Quality mode, limited to academic papers or chosen websites.
Add a local AI search endpoint to your own app through Vane's search API.
Pair it with a local Ollama model to run AI search without sending queries to a cloud LLM.

Get started

Run the Docker command, which includes SearxNG, then open http://localhost:3000. In the setup screen, add your API keys or point it at your local model server such as Ollama.

docker run -d -p 3000:3000 -v vane-data:/home/vane/data --name vane itzcrazykns1337/vane:latest

Commands are copied from the project's README. Read it on GitHub for the full setup.