— CASE 06 / ON-PREM AI · SILVER TOUCH
ON-PREM OCR
+ LLM.
Aadhaar, PAN, GST — data could not leave the client’s 250 GB server. Hybrid OCR + on-prem LLM with zero external API calls.
~90%Entity-extraction accuracy
200+Page PDFs handled
0External API calls
250GBOn-prem server · air-gapped
— THE BRIEF
DATA CAN’T
LEAVE THE
SERVER.
BUILD IT
ANYWAY.
Hybrid OCR combining DocLink (character accuracy) and DocTR (structure), merged by Llama-3.1-Nemotron-70b hosted on-prem via vLLM. Entity extraction layered on top with markdown / JSON output.
The hard constraint: a dedicated 250 GB on-prem server, every model self-hosted, zero external API calls — Aadhaar, PAN and GST numbers never leave the building. Entity extraction held ~90% accuracy across documents of varying scan quality, emitted as markdown or JSON.
VLLMMILVUSDOCTRDOCLINKNEMOTRON 70BBGE-M3ON-PREM GPU
— WANT THE FULL STORY?