Make documents, wikis and file repositories searchable using AI – as a RAG system on your own infrastructure, so that corporate knowledge remains in-house.
The knowledge is there – it’s just scattered. RAG systems make it searchable without sending it to third-party clouds.
Documents from file repositories and wikis are migrated as embeddings to a vector database on NVMe volumes – the source files remain in S3.
A self-hosted language model generates responses with source citations – running on a dedicated GPU, accessible internally.
Contracts, manuals and project knowledge never leave your VPC – this is the key difference compared to cloud-based AI services.
Access rights from the source systems carry over into the search – everyone can only see what they are authorised to view.
Open-source components such as LangChain, Qdrant and vLLM integrate with the RAG stack on the centron infrastructure: an ingest pipeline in Kubernetes, a vector database on volumes, and inference on GPUs. The result: an in-house search and answer engine that the legal department can approve.
Customers typically implement this use case using these building blocks – which can be combined and expanded at any time.
Knowledge management with AI: Make corporate knowledge searchable via RAG – on your own infrastructure, GDPR-compliant, with data remaining in-house. The core component is Cloud GPU from €92.59 per month – billed by the hour, with no minimum contract term. This is supplemented, as required, by Volumes and S3 Object Storage. Hosting is provided in compliance with the GDPR in centron’s own certified to ISO 27001 on the basis of IT-Grundschutz data centres, which hold BSI C5:2020 Type 1 certification. New accounts receive a €200 starting credit.
| Building block | Price |
|---|---|
| Cloud GPU | from €92.59 per month |
| Volumes | from €0.05 per month |
| S3 Object Storage | from €5.00 per month |
Sign up and receive €200 credit at centron within your first 60 days.
This promotional offer applies to new accounts only. Available exclusively to businesses.