# Word boxes that hold up on handwriting > Tighter word boxes on handwritten and scanned pages, and confidence scores that stop flagging words that are fine. - Canonical: https://www.datalab.to/blog/word-boxes-handwriting - Published: 2026-10-20 - Authors: Vik Paruchuri Our new word-box model is live in the API. On real handwritten and scanned pages, 93% of words now get a usable box, up from 60%, and correctly read words are far less likely to be flagged as low confidence. [Word bounding boxes and confidence](/blog/word-bounding-boxes-and-confidence) already worked well on clean, printed pages. On handwriting and old scans they were much weaker. Boxes drifted onto the next word or the next line, and many correctly read words came back with low confidence, so review queues filled up with text that didn't need a second look. ## Before and after Side-by-side view of a handwritten 19th-century pension declaration about Revolutionary War service. On the left, the previous model's word boxes overlap each other and spill across lines, with several words marked orange or red. On the right, the new model draws one tight box around each word on every line, almost all green. _A Revolutionary War pension declaration from the U.S. National Archives (public domain). Left: the previous model. Right: the new model. Box color is word confidence: green 0.9 and up, orange 0.5 to 0.9, red below 0.5._ | Real handwritten and scanned pages | Previous model | New model | | -------------------------------------------------- | -------------- | --------- | | Words with a usable box (IoU ≥ 0.5) | 60% | 93% | | Words with a tight box (IoU ≥ 0.75) | 12% | 72% | | Words with a box in the wrong place (IoU < 0.1) | 12% | 6% | | Correctly read words scored below 0.5 confidence | 4.4% | 0.8% | _Measured with our production code on held-out handwritten and scanned pages, against reference word boxes. Latency is unchanged._ Scanned forms improve too. Here are two printed forms filled in by hand: Side-by-side view of a 1910 U.S. passport application, a printed form filled in by hand, with an embassy stamp and seal. On the left, the previous model's boxes pile up around the stamp and over the handwritten entries. On the right, the new model draws one box around each printed and handwritten word, mostly green, with a few handwritten words in orange or red. _A 1910 U.S. passport application (public domain, via [Wikimedia Commons](https://commons.wikimedia.org/wiki/File:Passport_application_of_Edward_Kirkpatrick_%281910%29.jpg)). Left: the previous model. Right: the new model. The previous model's boxes pile up around the stamp and the handwritten entries. The new model boxes each printed and handwritten word on its own, and a few handwritten words are still scored orange or red._ Side-by-side view of Cole Porter's World War I draft registration card, a printed card filled in by hand. On the left, the previous model draws large boxes that cover several handwritten entries at once, and a stack of boxes over the stamp in the corner. On the right, the new model draws one box around each handwritten entry and each printed label. _Cole Porter's World War I draft registration card, from the U.S. National Archives (public domain, via [Wikimedia Commons](https://commons.wikimedia.org/wiki/File:World_War_I_Draft_Registration_Card_for_Cole_Porter_-_NARA_-_641783.gif)). Left: the previous model. Right: the new model._ ## Confidence you can route on Low confidence should mean "look at this word". On handwriting, it often meant "this is handwriting": 4.4% of correctly read words scored below 0.5. With the new model that is 0.8%, so a review queue built on confidence holds far fewer words that were fine. Side-by-side view of a handwritten 19th-century marriage affidavit. On the left, overlapping word boxes cover the page. On the right, each word has its own box, with a few overlaps in the last lines above the signature. A few words are marked orange or red. _A marriage affidavit from the same pension files. The new model gives each word its own box. It is not perfect: a few boxes still overlap in the last lines above the signature, and a few words are still scored orange or red._ ## What to check There's no API change. Turn on `word_bboxes` as before, and boxes and confidence come back in the same `data-bbox` and `data-confidence` attributes. The new model is live in both our US and EU regions. Confidence values on handwritten and scanned pages will be higher than before, because far fewer correct words score low. Page confidence, which is driven by the lowest-scoring words on the page, rises too. If you tuned a review threshold on the old scores, check it against a sample of your own pages. ## Try it Run a handwritten letter or an old scan through the [playground](https://www.datalab.to/playground) with word boxes turned on. If your archive has pages where the boxes still miss, we'd like to see them: [support@datalab.to](mailto:support@datalab.to).