Release status

Where Ovrin stands today: implemented through v0.3, no release tagged, confidence weights provisional, and no accuracy figure published yet.

Ovrin is implemented and designed, but the upstream project is still pre-v1. That means the project is not yet considered stable enough for a general production guarantee.

What this means

  • the library is feature-complete through the documented scope
  • adapters and architecture are present
  • no release tag is yet established in the public upstream repo
  • accuracy and calibration are still being formed by evaluation and real corpus work

Why it matters

For document extraction systems, accuracy and confidence calibration are not optional. A serious library must not overstate certainty until it has evidence from the real world.