Your AI's knowledge base, in France
An ultra-fast graph + vector database powered by FalkorDB and operated by DINAO. The foundation for your GraphRAG and AI agents — vectors and knowledge hosted on our French servers.
What is FalkorDB?
FalkorDB is a high-performance graph database with a native vector index. A modernized successor to RedisGraph, it represents the graph via sparse adjacency matrices and leverages linear algebra (GraphBLAS library) to execute some of the fastest queries on the market.
It implements the Property Graph model and the OpenCypher query language, allowing you to combine graph traversal, vector search (HNSW), and full-text search in a single query. This makes it an ideal building block for GraphRAG, LLM knowledge graphs, recommendation engines, and AI assistants.
On the technical side, the core is written in C/C++, FalkorDB operates as a Redis module, and comes with Python, Java, Node.js, Go, Rust, and C# clients, an official GraphRAG SDK, and a web administration interface. Docker deployment is immediate. (SSPLv1 License.)
Host FalkorDB at DINAO
Resource tiers compatible with FalkorDB prerequisites (minimum 1 vCPU / 512 Mo / 2 Go). Hosted in France, fully managed.
- 1 dedicated vCPU
- 2 Go RAM
- 20 GB NVMe
- Daily backups
- Managed & monitored by DINAO
- 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
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…
Is FalkorDB an AI / LLM?
No. FalkorDB does not generate text: it is the knowledge and vector base that your AI queries. It stores the graph and embeddings used by your LLMs for RAG and GraphRAG.
What does 'GraphRAG' offer compared to classic RAG?
GraphRAG combines vector search (semantic similarity) with a knowledge graph: your AI leverages not only nearby passages but also the relationships between entities, for more accurate and explainable answers.
Where are my vectors and graph hosted?
On DINAO infrastructure in France, in one of the available datacenters. Your embeddings and knowledge base never leave the country.
How do I integrate FalkorDB into my AI application?
Via the OpenCypher language and official clients (Python, Java, Node.js, Go, Rust, C#), or directly with the GraphRAG SDK which interfaces with LLMs. FalkorDB operates as a Redis module, compatible with the associated ecosystem.
Can I change plans or export my data?
Yes. You can upgrade or downgrade at any time, and your graph and vectors remain exportable — no proprietary lock-in.
