Chat with your documents, privately
A document Q&A AI (RAG) powered by Kotaemon, installed and maintained by DINAO. Your files and questions stay on our French servers — in local mode, they never leave the premises.
What is Kotaemon?
Kotaemon is an open-source, streamlined, and customizable RAG (Retrieval Augmented Generation) interface designed to dialogue with your documents. Built for both end-users and developers, it offers a Q&A experience where every answer is based on your own files and cites its sources.
The engine combines hybrid search (full-text + vector) with re-ranking, and supports multimodal content including figures, tables, and OCR. An integrated PDF viewer highlights cited passages and their relevance scores. Kotaemon is multi-provider: it works with local models (Ollama, llama-cpp-python) as well as external APIs (OpenAI, Azure, Cohere).
Its extensible architecture allows you to build custom RAG pipelines and advanced reasoning methods, making it a solid foundation for enterprise document AI — deployed in containers and scaled by DINAO according to your models.
Host Kotaemon at DINAO
Resource tiers compatible with Kotaemon prerequisites (minimum 2 vCPU / 4 Go / 10 Go (image main-lite)). 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 may also like…
Other applications managed by DINAO in the same category, hosted in France and GDPR compliant.
You might be wondering…
Are my documents used to train models?
No. In local model mode, your documents and questions remain private on your instance and are never shared or used for training. This is the core benefit of sovereign RAG.
What document formats can I query?
Kotaemon indexes PDFs, office documents, and text, with multimodal support (OCR, figures, tables). You query everything in natural language and get sourced answers.
Can I use my own models?
Yes. You can choose from locally hosted open-weight models (Ollama, llama-cpp) or connect an external API (OpenAI, Cohere…) with your own key.
Do I need a GPU?
Not necessarily. Small models (3–7B) run on CPU. For larger models or high throughput, a GPU is recommended: it is available as an option on Team and Enterprise tiers.
Where is the data hosted?
On DINAO infrastructure in France, in one of the available data centers. In local mode, your data never leaves the country.
