# LayoutLMv3 Inference

**URL:** <https://discuss.huggingface.co/t/layoutlmv3-inference/27118>\
**Category:** Intermediate\
**Created:** [December 2, 2022, 11:39am UTC](https://discuss.huggingface.co/t/layoutlmv3-inference/27118 "2022-12-02T11:39:11Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![WaterKnight](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/waterknight/32/33976_2.png) [@WaterKnight](https://discuss.huggingface.co/u/WaterKnight)\
**Post date:** [December 2, 2022, 11:39am UTC](https://discuss.huggingface.co/t/layoutlmv3-inference/27118/1 "2022-12-02T11:39:11Z")

</div>

Hi, I have seen the tutorial from @nielsr [Transformers-Tutorials/LayoutLMv3 at master · NielsRogge/Transformers-Tutorials · GitHub](https://github.com/NielsRogge/Transformers-Tutorials/tree/master/LayoutLMv3)

However, I wanted to know how to get the words of each box, because in his example he is just using it to draw boxes

---

<div class="post-metadata">

**Author:** ![rishisriv-bh](https://avatars.discourse-cdn.com/v4/letter/r/df788c/32.png) [@rishisriv-bh](https://discuss.huggingface.co/u/rishisriv-bh)\
**Post date:** [March 9, 2024, 4:29pm UTC](https://discuss.huggingface.co/t/layoutlmv3-inference/27118/2 "2024-03-09T16:29:38Z")

</div>

Did you have any luck in finding how to get the words of each box?

---

<div class="post-metadata">

**Author:** ![nielsr](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/nielsr/32/39617_2.png) [@nielsr](https://discuss.huggingface.co/u/nielsr)\
**Post date:** [March 11, 2024, 9:15pm UTC](https://discuss.huggingface.co/t/layoutlmv3-inference/27118/3 "2024-03-11T21:15:27Z")

</div>

Could you clarify? The OCR engine gives you the boxes along with their words.