An AI search engine, private and hosted in France
A sovereign alternative to Perplexity powered by Perplexica, installed and maintained by DINAO. Your searches stay on our French servers — never profiled by a third-party service.
What is Perplexica?
Perplexica is an AI-augmented search engine, open-source, designed as an alternative to Perplexity AI. Rather than returning a simple list of links, it queries the web via the privacy-respecting SearXNG metasearch engine, then leverages a large language model to synthesize a clear and sourced answer, citing the pages used.
It offers several specialized search modes — general web, academic, videos, writing research — and uses embedding and similarity techniques to refine relevance. On the model side, Perplexica works with a local open-weight LLM via Ollama, or connects to an external API (OpenAI, Anthropic, Mistral…) with your own key.
Managed by DINAO and connected to a local model, Perplexica becomes a truly private and sovereign research assistant: your queries, your monitoring topics, and your results remain confidential, on our French servers, safe from public profiling.
Host Perplexica at DINAO
Resource tiers compatible with Perplexica 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 is this different from Perplexity?
The principle is the same — an AI-augmented search that cites its sources — but here everything runs on your DINAO instance in France. Your queries are neither stored nor profiled by a third-party service, and you choose your model.
Can I use my own models?
Yes. Perplexica connects to a locally hosted open-weight LLM (Ollama) or to an external API with your key. For sovereignty, prefer the local mode.
Where are the data hosted?
On DINAO infrastructure in France, in one of the available datacenters. Your searches do not leave the territory.
Do I need technical skills?
No for usage: the interface is ready to use. DINAO handles installation, the metasearch engine, the LLM, and updates. A technical team can leverage the API and SSO.
Are my searches used to train a model?
No. In local mode, your queries and results remain private on your instance and are never shared or used to train a third-party model.
