Your AI research assistant, private and in France
A deep research assistant powered by Local Deep Research, installed and maintained by DINAO. Sourced reports from the web, arXiv, PubMed, and your documents — everything stays on our French servers.
What is Local Deep Research?
Local Deep Research (LDR) is an open-source deep research assistant designed to run entirely locally so that no data leaves your infrastructure. It performs systematic research: it breaks down complex questions, queries multiple sources in parallel, and then produces comprehensive and properly cited reports.
The tool queries more than ten sources — the web, arXiv, PubMed, Wikipedia, GitHub — as well as your private documents, iterates on its findings, and synthesizes everything into a structured report. It supports local LLMs (Ollama, llama.cpp) as well as cloud models, and achieves excellent scores on question-answering benchmarks.
LDR also provides an MCP server (Model Context Protocol) that exposes its research capabilities to assistants like Claude Desktop or Claude Code. With encrypted storage and local operation, it is an ideal option for privacy-respecting AI research.
Host Local Deep Research at DINAO
Resource tiers compatible with Local Deep Research prerequisites (minimum 2 vCPU / 4 Go / 10 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…
What is "deep research"?
Rather than a simple answer, the assistant breaks down your question, launches multiple parallel searches across various sources, iterates on what it finds, and then writes a structured and cited report — much like a document analyst.
Do my searches remain private?
In local mode, yes: your questions, documents, and reports are encrypted and stay on your DINAO instance in France. Nothing is transmitted to a third-party service.
Which sources are queried?
More than ten: the web, arXiv, PubMed, Wikipedia, GitHub, and especially your own indexed private documents, all combined in a single search.
Can I use my own models?
Yes. The tool works with local LLMs (Ollama, llama.cpp) or external APIs. For sovereignty, prioritize local models.
Can it be connected to Claude or another assistant?
Yes. Local Deep Research exposes an MCP server that makes its research capabilities available in assistants like Claude Desktop or Claude Code.
