Open-source · AGPL-3.0
Enterprise RAG.
Self‑hosted. EU‑sovereign.
The open-source RAG platform for organizations that cannot send their data to American clouds. One-command setup, on-premise or air-gapped.
Built on European open-source
Sovereignty isn't a slogan — it's a supply chain.
Every line we depend on is open-source, auditable, and developed inside the European Union. No US cloud in the request path.
+ more EU open-source projects we contribute to
Why I3K RAG Enterprise
Built for organizations that can't outsource their data.
Most RAG platforms assume you're happy sending your documents to a US cloud. We assume you're not — and we engineer accordingly.
EU-sovereign by design
Servers, data, models and the engineering team are all in the European Union. No transatlantic data transfers, no Schrems II exposure.
Backup & restore enterprise-grade
Full system backup with 70+ cloud providers via rclone. S3, MEGA, Google Drive, OneDrive, Dropbox, WebDAV/Nextcloud, FTP, SFTP, Backblaze B2, pCloud. Cron scheduling, retention policies, zero-downtime.
Runs on the hardware you have
NVIDIA CUDA, AMD ROCm, or CPU-only. No vendor lock-in. Choose during setup.
Multilingual out of the box
BAAI/bge-m3 embeddings support 29 languages out of the box. No per-language fine-tuning.
On-premise first
One-command install on Ubuntu 20.04+. Runs fully air-gapped on your hardware. No outbound calls to third-party APIs.
GDPR & EU AI Act ready
JWT auth, RBAC with 3 roles, audit log and retention policies. AGPL-3.0 source — auditable end to end.
How it works
The RAG pipeline in 4 steps.
Ingest, embed, retrieve, generate. Every stage runs locally. No data ever leaves your infrastructure.
- 01
Ingest
Upload via web UI or API. Apache Tika + Tesseract extract text from PDF, DOCX, PPTX, XLSX, ODT, RTF, HTML, XML and scanned documents (OCR).
Tika · OCR
- 02
Embed & store
Documents are chunked and embedded with BAAI/bge-m3 (29 languages). Vectors stored in Qdrant with metadata for RBAC filtering.
bge-m3 · Qdrant
- 03
Retrieve
Our in-house retrieval orchestrator runs semantic retrieval from Qdrant. Configurable relevance threshold and top-K. Role-based filtering at retrieval layer.
I3K orchestrator
- 04
Generate
Retrieved chunks passed to EuLLM (default) or compatible LLM for grounded answer generation. Default models: Qwen3:14b-q4_K_M, Mistral 7B Q4. Fully local, zero external calls.
EuLLM · Mistral 7B
29 Languages
70+ Backup destinations
10,000+ Documents per node
100% Local
Compare
I3K RAG Enterprise vs the alternatives.
We've tried to be fair. If we've got something wrong about a competitor, tell us.
| Capability | I3K RAG Enterprise | Onyx | Glean | Cohere North |
|---|---|---|---|---|
| Open-source core | Yes | Yes | No | No |
| EU-sovereign LLM engine included (EuLLM) | Yes | No | No | No |
| 100% self-hosted | Yes | Yes | Hybrid | No |
| Air-gapped deployment | Yes | Partial | No | No |
| EU-resident development & data | Yes | No | No | No |
| Built-in backup (70+ cloud providers) | Yes | No | N/A | N/A |
| Multi-GPU support (NVIDIA/AMD/CPU) | Yes | Limited | N/A | N/A |
| Multilingual (29 languages, bge-m3) | Yes | EN-first | EN-first | Yes |
| GDPR + EU AI Act tooling | Yes | DIY | Add-on | Add-on |
| No vendor lock-in | Yes | Yes | No | No |
Comparison based on publicly available documentation as of 2026. Vendor capabilities evolve — verify against the latest releases.
Ready to run RAG on your own infrastructure?
Start with the open-source Community edition, or talk to us about Pro with structured extraction, SSO, audit log and SLA.