Vane
Vane is a privacy-focused AI answering engine that runs on your own hardware. It combines web search with support for local LLMs (Ollama) and cloud providers (OpenAI, Claude, Groq, Gemini) to deliver accurate answers with cited sources — keeping your searches completely private.
Vane is a privacy-focused AI answering engine that runs on your own hardware. It combines web search with support for local LLMs (Ollama) and cloud providers (OpenAI, Claude, Groq, Gemini) to deliver accurate answers with cited sources — keeping your searches completely private.
MrBandiサービス
Vane — Privacy-Focused AI Answering Engine
Vane is an AI-powered answering engine that runs entirely on your own hardware. It uses SearxNG for private web search and supports multiple AI providers including local LLMs via Ollama, OpenAI, Anthropic Claude, Google Gemini, and Groq.
Key Features
- Privacy-first — All searches stay on your infrastructure
- Multiple AI providers — Ollama, OpenAI, Claude, Gemini, Groq, and more
- Smart search modes — Speed, Balanced, and Quality modes
- Cited sources — Answers include references to source materials
- Web, image & video search — Powered by SearxNG
- File uploads — Ask questions about your documents
Deployment Details
- Image:
itzcrazykns1337/vane:latest(includes bundled SearxNG) - Port:
3000(HTTP) - Persistent Storage: Mounted at
/home/vane/data - Domain: Bind your custom domain with
PUBLIC_DOMAIN
After deployment, visit your ${ZEABUR_WEB_URL} to configure AI providers and settings in the setup screen.