Your documents that answer, hosted in France
A deep document understanding RAG engine powered by RAGFlow, installed and maintained by DINAO. Ask your questions, get sourced answers based on your data — all on our French servers, never on an American cloud.
What is RAGFlow?
RAGFlow, published by InfiniFlow, is a leading open source RAG engine, based on deep document understanding. Coupled with a language model, it provides reliable and sourced answers with citations from data in complex formats : Word, slides, Excel, text, images, scanned copies, structured data, or web pages.
Its strength lies in deep learning-based parsers that extract knowledge from unstructured document units, intelligent chunking, and chunk visualization allowing human intervention. Key references and traceable citations ensure well-founded answers. A system of visual agents (Canvas) allows building multi-step AI applications with persistent memory, web search, and Deep Research-style reasoning.
RAGFlow combines cutting-edge RAG and agent capabilities to offer a superior context layer for LLMs. Managed by DINAO and connected to a local model, it makes your documents speak while keeping 100% of your data in France.
Host RAGFlow at DINAO
Resource tiers compatible with RAGFlow prerequisites (minimum 4 cœurs / 16 Go / 50 Go). Hosted in France, fully managed.
- 8 dedicated vCPU
- 16 Go RAM
- 160 GB NVMe
- Daily backups
- Managed & monitored by DINAO
3 hidden tier(s) (insufficient resources for this app) : Découverte, Standard, Performance
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 'deep document understanding'?
RAGFlow uses deep learning-based parsers to analyze the structure of complex documents (tables, layouts, scanned images). It extracts reliable knowledge, where naive chunking would fail.
Do my documents go to OpenAI?
Not in local mode. Connected to an open-weight model hosted on your DINAO instance, your documents and conversations remain in France. This is the truly sovereign mode (see the sovereignty box).
Are the answers reliable?
RAGFlow bases each answer on verifiable citations from your documents and allows visualizing the chunking to control and correct. This significantly reduces hallucinated answers.
Can I create AI workflows?
Yes. The visual agent system (Canvas) allows building multi-step AI applications with persistent memory, web search, and Deep Research-style reasoning.
Do I need technical skills?
DINAO handles deployment, security, and updates. The knowledge base management interface is accessible; your technical teams can additionally use the API and agents.
