Docen OCR 2
Our recognition model gets better at faint scans, handwriting, and dense multilingual pages.
THE DOCEN TEAM·Mar 24, 2026·5 MIN READ
Our recognition model gets better at faint scans, handwriting, and dense multilingual pages.
What's new
We built this release around Docen OCR, with a focus on the messy, real-world inputs that separate a good result from a usable one.
We reworked how the model handles edge layouts and low-quality scans, then measured the difference against our own labeled benchmarks before release.
- Improved handling of faint scans.
- Improved handling of handwriting.
- Improved handling of multilingual pages.
By the numbers
+7%
accuracy on hard pages
−34%
median latency
95
languages supported
Getting started
The update is live on the managed API and available for self-hosted deployments. Existing integrations pick it up with no code changes — the output format is unchanged.
Want to see it on your own documents? Open the playground or reach out — we're happy to run a sample with you.
faint scanshandwritingmultilingual pages