Skip to content
RepoQuarry

Vane (formerly Perplexica) ItzCrazyKns/Vane

Self-hosted AI answering engine that searches the web through SearxNG and answers with cited sources using local or cloud LLMs.

Vane (formerly Perplexica) at a glance

Checked 15 Oct 2026

Vane, the renamed Perplexica (GitHub redirects ItzCrazyKns/Perplexica to ItzCrazyKns/Vane), is a self-hosted Perplexity-style answer engine. It searches via a bundled SearxNG, answers with citations, offers speed, balanced and quality modes, and answers questions about uploaded files. Models come from Ollama or Lemonade locally, or from OpenAI, Claude, Gemini and Groq. One Docker command runs it. MIT licensed; ships no weights, so the chosen model's licence and quality matter. Hardware needs are not documented.

Cite this pageRepoQuarry. “Vane (formerly Perplexica).” Checked 15 Oct 2026. https://repoquarry.dev/projects/vane

How much of Perplexity Pro does it replace?

2 of 3 jobs. Replaces the cited web-answer interface; answer quality depends on the model you run. No premium data sources or Computer features.

Partial: Web search answers Partial replacement
No: Chat assistant Not covered
Partial: Chat with your documents Partial replacement

For chat assistant, see other Perplexity Pro alternatives. You may also need to build an interface, an API server and hosting yourself.

Facts, with sources

Every answer links to where we found it. We read the documentation; we did not install it.

Report a correction
QuestionAnswerSourceChecked
Runs on my machine? Confirmed: Yes, self-hosted via Docker or Node.js Vane is a privacy-focused AI answering engine that runs entirely on your own hardware. github.com · Vane#readme ↗ 1d ago
Code licence Confirmed: MIT github.com · LICENSE ↗ 1d ago
Model weights licence Confirmed: No bundled weights; licence depends on the model you connect Use local LLMs through Ollama or connect to OpenAI, Anthropic Claude, Google Gemini, Groq, and more. github.com · Vane#-features ↗ 1d ago
Commercial use Confirmed: Yes, MIT code; check the licence or terms of the model you use github.com · LICENSE ↗ 1d ago
Paid services needed? Partial: Works with local Ollama + bundled SearxNG; cloud LLM APIs optional The image includes both Vane and SearxNG, so no additional setup is required. github.com · Vane#getting-started-with-docker-recommended ↗ 1d ago
Hardware Unknown: Not documented; depends on the model you run Know the answer? —
Operating systems Confirmed: Docker on Windows, macOS and Linux (per-OS Ollama setup documented) Windows: Use http://host.docker.internal:11434 ... Mac: Use http://host.docker.internal:11434 ... Linux: Use http://<private_ip_of_host>:11434 github.com · Vane#ollama-connection-errors ↗ 1d ago
Setup effort Confirmed: Medium; one Docker command (SearxNG bundled), or npm build + own SearxNG There are mainly 2 ways of installing Vane - With Docker, Without Docker. Using Docker is highly recommended. github.com · Vane#installation ↗ 1d ago

Get the weekly top 5 and what’s new, every Monday

We’ll email you once a week. Unsubscribe any time. We never share your email.

Cookie preferences

Choose what we may store in your browser. You can change this any time from Cookie settings in the footer.

Essential

Security, and remembering this choice. rq-consent · 12 months

Always on