Regulated by & built for
Certified by
Funded by
c:node is the EU-sovereign AI operator across GPT, Claude, Gemini & open models — and NENA lays the evidenced company knowledge on top, with a source instead of a guess.
Connect your sources — SAP, HubSpot, Microsoft 365 & more — and feed NENA, the knowledge store that grows with every tenant. c:node pulls data via API from the systems you already use. Missing yours? We’ll wire it up.
+ any system with an API
Launch your own AI under your logo — with no AI team, without a line of code. c:node ships the finished app, hosted & maintained; NENA the trained, evidenced intelligence behind it. One platform, any number of brands — productive from day one.
NENA is the memory of c:node: it remembers your company knowledge and answers questions about it — every answer with a source, nothing guessed, nothing forgotten.
You steer, NENA thinks along, the agents act. How NENA works and why it doesn’t hallucinate — see the scientific background.
Scientific backgroundc:node-Shell is open source on GitHub (AGPL-3.0): the full app, the graph runtime and the connectors. Clone it, docker compose up, run it locally — on your infrastructure, air-gapped if needed.
docker compose up — local & air-gappedOpen Core = the shell: UI, graph runtime & connectors. The trained NENA intelligence is not included — it comes with Cloud, Business & Whitelabel.
View on GitHubc:node is model-agnostic: pick the best model for the job — GPT-4o, Claude, Gemini via API or open models on-prem (Ollama: qwen2.5, llama3, mistral). c:node orchestrates the models and the agents. On top sits the real intelligence core: NENA — neuro-symbolic, above every language model. It thinks proactively in context, makes every answer traceable (a source, not a black box) and runs sovereignly on-prem. One core across all models — that’s NENA.
qwen2.5:7b
llama3.1:8b
mistral:7b
qwen2.5:7b
llama3.1:8b
mistral:7b
We bring NENA on-prem — then it learns your knowledge and grows with your operation.
Faster, more integrated, collaborative: your daily knowledge and workflows in one place.
Answers from your real knowledge in seconds — instead of searching ten systems.
c:node sits above your tools and connects what otherwise stays separate.
Shared memory for the team — everyone asks, everyone sees the source.
c:node runs right where you are — on your server, GDPR-compliant, online and offline. Your data stays exactly where it belongs: with you. And leaves your building only if you want it to.
Entirely in your data centre.
Without any internet connection.
Hosted in Europe, if you prefer.
No. The intelligence NENA works neuro-symbolically (rules and patterns) on a knowledge graph and makes every statement traceable, with a source — instead of generating freely.
Yes. c:node runs on-premise, air-gapped or in the EU cloud — online and offline, GDPR-compliant and EU-AI-Act-ready. Your data stays with you.
c:node connects to your stack — including SAP, HubSpot, Microsoft 365, DATEV, Nextcloud — as well as PDF, CSV, APIs and registers. More connectors on an ongoing basis.
NENA is the intelligence behind c:node: Neuro-symbolic · Evidence · Network · Agent. c:node steers, NENA thinks, the agents act — intelligence in any stack, online and offline.
Yes. c:node is fully whitelabel-capable — the same platform, your logo, your brand, for team- and org-wide rollout.
In a call we walk you through a short test run on your real data — and show how c:node answers with a source, not a guess.
