What it sells
Self-hosted private AI stack on a single server: open-weight model, optimized inference runtime, OpenAI/Anthropic-compatible APIs and enterprise controls.
Business model breakdown
Buyer: who owns the budget?
Enterprise IT/AI leads with Finance, Legal and Procurement sign-off (site says the model is predictable for Finance, clean for Legal, simple for Procurement).
Trigger: what makes them buy now?
Data cannot leave the premises (EU AI Act, air-gapped needs) and per-token API costs are unpredictable; flat-rate license per node removes token metering.
Workflow: what inputs become what output?
Deploy a prebuilt stack on own GPU server, connect tools such as Claude Code, Open WebUI, LangChain; run agents in off-hours.
Economics: what it saves and how it charges
Flat-rate license per server per year, no per-token billing; claims ~0.33x (Max) and ~0.11x (Compact) cost per task versus a frontier API.
Distribution: how it reaches customers
Try it, live in 24h; partners page.
Funding history and coverage
| Round date | Stage | Amount | Investors | Source |
|---|---|---|---|---|
| 2026-08-31 | — | 800 thousand euro | — | news.google.com |
Profiles summarize public reporting and company websites. Business-model notes are MaxBC's reading of that material, not confirmed facts; unknowns are left blank rather than guessed. Updated 2026-10-12.