BKS-Lab

We are a small team in Cologne that builds AI systems which run where our customers' data is allowed to stay: on their own servers, or with a provider under EU jurisdiction. Most of our work is making confidential company knowledge searchable and usable for automation, with open models we can host ourselves.

Before we recommend a local model, we measure it. Our test bench runs open models on one RTX 4090 and compares them with hosted services on the same tasks: German retrieval for RAG, German speech to text, invoice fields from rendered page images, routing and extraction. Code, cases and raw rows are on GitHub, the result tables are here.

• Results: lab-bench-results (dataset) and lab-bench (Space)
• Invoice fields from page images: xrechnung-invoice-fields (34 XRechnung test invoices with gold fields and our models' outputs) and its viewer
• Models we measured, with our numbers: Measured on one RTX 4090 (collection)
• Test bench on GitHub: bks-lab/lab-bench
• Write-ups: bks-lab.com/en/labor/test-bench
• Our agent setup, open source: bks-lab/open-bridge
• Why sovereign AI: bks-lab.com/en/souveraene-ki

Deutsch. Wir bauen KI-Systeme, die dort laufen, wo die Daten bleiben dürfen. Welches offene Modell eine Aufgabe trägt, messen wir vorher auf eigener Hardware und legen die Messungen offen: bks-lab.com/labor/pruefstand.