Here's a direct link to the code: https://github.com/caesarnine/llm-experiments/blob/main/3_structured_ocr_with_gpt_vision/pdf_parser.py https://github.com/caes

Here's a direct link to the code: https://github.com/caesarnine/llm-experiments/blob/main/3_structured_ocr_with_gpt_vision/pdf_parser.py https://github.com/caesarnine/llm-experiments/blob/main/3_st... Just a Pydantic model to define an extraction schema + a simple prompt works reasonably well.