Your private AI search engine, hosted in France
An AI answer engine that searches the web and cites its sources powered by Vane, installed and maintained by DINAO. Your queries stay on our French servers — they never go to OpenAI.
What is Vane?
Vane (formerly Perplexica) is a self-hosted AI search engine: a private alternative to services like Perplexity. It combines a private metasearch (via SearXNG) and a language model of your choice to answer your questions in natural language.
Its key feature is that it cites its sources for every statement: instead of having to trust the model, you can verify the information yourself. Vane also allows you to upload files, maintain a persistent search history, and use multiple search modes.
Regarding models, Vane works with a local LLM (served by Ollama or llama.cpp) or an external API. Locally, usage is entirely free and private. The application is free, open source, and deploys easily via Docker on your own hardware for total privacy of your searches.
Host Vane at DINAO
Resource tiers compatible with Vane prerequisites (minimum 2 vCPU / 2 Go / 5 Go). Hosted in France, fully managed.
- 2 dedicated vCPU
- 4 Go RAM
- 40 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 4 dedicated vCPU
- 8 Go RAM
- 80 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 8 dedicated vCPU
- 16 Go RAM
- 160 GB NVMe
- Daily backups
- Managed & monitored by DINAO
1 hidden tier(s) (insufficient resources for this app) : Découverte
This application uses AI
The container hosts the application, not the AI engine (which requires dedicated GPUs) : inference is handled externally, with your own provider key. Prioritize a sovereign engine — Mistral AI, NumSpot, Scaleway, or OVHcloud AI Endpoints (France, GDPR) ; an international provider (OpenAI, Anthropic…) only if a specific capability requires it. Inference subscriptions are not included in hosting.
Technical details
You might be wondering…
How does Vane differ from ChatGPT or Perplexity?
Vane is self-hosted: it combines private metasearch (SearXNG) and your own models. In local mode, your queries are never sent to OpenAI or a third-party service.
Are the answers reliable?
Vane cites its sources for every answer, allowing you to verify the information yourself instead of blindly trusting the model.
Can I use my own models?
Yes. You connect a local open-weight model served by Ollama, or an external API with your own key. Local inference remains free and private.
Where is the data hosted?
On DINAO infrastructure in France, in one of the available datacenters. In local mode, your queries and history never leave the territory.
Can I change plans or export my data?
Yes. You can upgrade or downgrade at any time, and your history remains exportable — no proprietary lock-in.
