from Hacker News

Structured OCR with GPT Vision

by binalpatel on 12/11/23, 1:59 PM with 1 comments

• by binalpatel on 12/11/23, 2:22 PM

  Here's a direct link to the code: 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.