# Handwritten OCR w/ confidence scores

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhandwritten-ocr-w-confidence-scores%2F143476%3E%5C
**Category:** Beginners\
**Created:** [February 28, 2025, 7:21pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhandwritten-ocr-w-confidence-scores%2F143476 "2025-02-28T19:21:51Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![riccardodemaria](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/riccardodemaria/32/39915_2.png) [@riccardodemaria](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Friccardodemaria)%5C
**Post date:** [February 28, 2025, 7:21pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhandwritten-ocr-w-confidence-scores%2F143476%2F1 "2025-02-28T19:21:51Z")

</div>

Hello everyone,

I am currently looking for suggestions to implement a handwritten unstructured invoice parsing pipeline.

What open-source models do you recommend for handwritten ocr/parsing?  
I have tried EaysOCR, Qwen, Intern-MPO, LayoutLM but they all seem to achieve poor results with handwritten invoices.

The idea is to find an open-source alternative to Textract OCR, so that I can fine-tune it when Textract performs poorly.

Thank you!

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [March 1, 2025, 6:20am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhandwritten-ocr-w-confidence-scores%2F143476%2F2 "2025-03-01T06:20:48Z")

</div>

It seems that some people are training Qwen 2 and Llama 3 Vision for handwritten characters. These models are high-performance, so they might work. Also, Microsoft’s Florence 2 and Google’s Paligemma 2 are also high-performance. I don’t know if they are suitable for handwritten character OCR…

> **[alxfgh/Doctor-Handwriting-Recoginizer-Qwen2-VL · Hugging Face](https://huggingface.co/alxfgh/Doctor-Handwriting-Recoginizer-Qwen2-VL)**
>
> We’re on a journey to advance and democratize artificial intelligence through open source and open science.

> **[Qwen2.5-VL | Try Now the Revolutionary Vision-Language Model](https://qwen-ai.com/2-5-vl/)**
>
> In the rapidly evolving world of AI, Qwen2.5-VL stands out as a monumental leap forward in vision-language technology. Building on the success of earlier Qwen