The vector database for your AI applications
A vector database powered by Weaviate, installed and maintained by DINAO. Semantic search and RAG on your data — hosted in France, not in an American cloud.
What is Weaviate?
Weaviate is an open-source vector database and cloud-native. It stores both objects and their vectors, combining vector (semantic) search, hybrid search, and structured filtering : it is a core infrastructure component for AI applications (RAG, semantic search).
Written in Go with high-performance HNSW indexing, Weaviate exposes REST, gRPC, and GraphQL APIs. Its vectorizer and generative modules connect to embedding and generation models — local (Ollama) or remote (OpenAI, Cohere, HuggingFace…) — to vectorize your data on import and feed your RAG pipelines.
Licensed under BSD-3-Clause and deployable in Docker / Kubernetes, Weaviate is fully managed by DINAO : installation, HTTPS, backups, and updates are handled, providing your AI applications with a high-performance vector memory hosted in France.
Host Weaviate at DINAO
Resource tiers compatible with Weaviate prerequisites (minimum 2 vCPU / 2 Go / 5 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 a vector database used for?
It stores 'embeddings' (numerical representations of your text, images, etc.) and enables semantic similarity search. It is the essential component for RAG and intelligent document search.
Do my data go to an AI provider?
Not in sovereign mode. If you connect Weaviate to a local embedding model (e.g., Ollama on your DINAO instance), everything stays in France. See the sovereignty note below.
Which models can I connect?
Weaviate integrates with many vectorizers: local models (Ollama) for sovereign use, or external APIs (OpenAI, Cohere, HuggingFace…) if you choose to.
What volume can Weaviate handle?
From a few thousand to several million vectors depending on the plan. Thanks to HNSW indexing and cloud-native architecture, latency remains controlled at scale.
Are technical skills required?
Application integration is required (via REST, gRPC, GraphQL, or Python/JS clients). DINAO handles infrastructure installation, security, and updates.
