Your AI search engine, hosted in France
A production-ready RAG engine powered by R2R (SciPhi), installed and maintained by DINAO. Query your own documents with AI — your data stays on our French servers, never on an American cloud.
What is R2R?
R2R ("RAG to Riches"), published by SciPhi, is an advanced AI retrieval system and a production-ready RAG engine. Built around a containerized RESTful API, it allows you to ingest multimodal content (text, PDF, JSON, images, audio…), perform hybrid search, and build knowledge graphs, all with full document and user management.
R2R combines semantic and keyword search (reciprocal rank fusion), a Graph RAG that automatically extracts relationships between your content, and an observability layer to analyze the performance of your retrieval pipeline. Authentication, role management, and an admin dashboard are integrated.
Fully open source (MIT license), R2R deploys via Docker. Managed by DINAO and connected to a local model, it offers you AI-augmented search on your own documents while keeping 100% of your data in France.
Host R2R at DINAO
Resource tiers compatible with R2R prerequisites (minimum 4 cœurs / 8 Go / 20 Go). Hosted in France, fully managed.
- 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
2 hidden tier(s) (insufficient resources for this app) : Découverte, Standard
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 may also like…
Other applications managed by DINAO in the same category, hosted in France and GDPR compliant.
You might be wondering…
What is a RAG engine?
RAG (Retrieval-Augmented Generation) combines search within your documents and generation by a language model: AI answers your questions based on your own content, with verifiable citations, rather than relying solely on its general knowledge.
Do my documents go to OpenAI?
Not in local mode. Connected to an open-weight model hosted on your DINAO instance, your documents and queries remain in France. This is the truly sovereign mode (see the sovereignty box).
What types of files can I ingest?
R2R is multimodal: text, PDF, JSON, images, audio, and more. It extracts the meaning, builds a hybrid index, and a queryable knowledge graph.
Can I integrate it into my applications?
Yes. R2R exposes a containerized RESTful API with authentication and user management: you integrate AI search into your business applications with just a few calls.
Do I need technical skills?
DINAO handles deployment, security, and updates. Usage via the dashboard is accessible; your developers can additionally leverage the API and observability.
