←back to thread

Run structured extraction on documents/images locally with Ollama and Pydantic

(github.com)
166 points EarlyOom | 1 comments | 20 Feb 25 01:54 UTC | HN request time: 0.219s | source
Show context
jbmsf ◴[20 Feb 25 03:10 UTC] No.43110666[source]▶
Interesting. We're using a SAAS solution for document extraction right now. I don't know if it's in our interest to build out more but I do like the idea of keeping extraction local.
replies(2): >>43110817 #>>43111598 #
fzysingularity ◴[20 Feb 25 05:59 UTC] No.43111598[source]▶
Cool, what types of documents do you currently handle? We could share some of our learnings/schemas here too.
replies(2): >>43111792 #>>43124212 #
andrewinardeer ◴[20 Feb 25 06:39 UTC] No.43111792[source]▶
Different commenter; Here I'm extracting data from commerical invoices, POs and bills of lading.
replies(1): >>43111865 #
1. fzysingularity ◴[20 Feb 25 06:53 UTC] No.43111865[source]▶
Ah cool, care to share a few examples? We can probably add those schemas in the next few days if there's enough folks who could benefit from this. A basic invoice schema is already there: https://github.com/vlm-run/vlmrun-hub/blob/main/vlmrun/hub/s...

You can see some of the qualitative results on GPT4o, Gemini, Llama 3.2 11B, Phi-4 here: https://github.com/vlm-run/vlmrun-hub?tab=readme-ov-file#-qu...