# Gemini 2.0 and PDF OCR Fine-tuning

**URL:** <https://discuss.ai.google.dev/t/gemini-2-0-and-pdf-ocr-fine-tuning/73178>\
**Category:** Google AI Studio\
**Tags:** gemini-flash, ai-studio, fine-tuning\
**Created:** [March 18, 2025, 3:45pm UTC](https://discuss.ai.google.dev/t/gemini-2-0-and-pdf-ocr-fine-tuning/73178 "2025-03-18T15:45:40Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Jesse\_Merrigan](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/jesse_merrigan/32/23958_2.png) [@Jesse\_Merrigan](https://discuss.ai.google.dev/u/Jesse_Merrigan)\
**Post date:** [March 18, 2025, 3:45pm UTC](https://discuss.ai.google.dev/t/gemini-2-0-and-pdf-ocr-fine-tuning/73178/1 "2025-03-18T15:45:40Z")

</div>

Hi everyone, I’m currently experimenting with Gemini 2.0 Flash for OCR on PDF documents. I have a relatively long prompt that instructs how I’d like the model to process the document (including deciphering relevant text and extracting handwritten text, formatting in a specific way, and how it should use confidence scores to judge when to correct OCR-based inaccuracies vs. retaining genuine spelling errors from the original handwritten text).

However, I feel I could improve the one-shot outputs I’m getting if I could include/attach .pdf documents in user inputs for fine-tuning jobs. Does anyone know if this is possible anytime soon, or if there’s an equivalent workaround?

Cheers!

---

<div class="post-metadata">

**Author:** ![GUNAND\_MAYANGLAMBAM](https://d1dlmcr85iqnpo.cloudfront.net/user_avatar/discuss.ai.google.dev/gunand_mayanglambam/32/4964_2.png) [@GUNAND\_MAYANGLAMBAM](https://discuss.ai.google.dev/u/GUNAND_MAYANGLAMBAM)\
**Post date:** [June 12, 2025, 10:14am UTC](https://discuss.ai.google.dev/t/gemini-2-0-and-pdf-ocr-fine-tuning/73178/3 "2025-06-12T10:14:54Z")

</div>

Hey @Jesse_Merrigan

Unfortunately, direct fine-tuning of Gemini 2.0 Flash with PDF files isn’t currently possible.