Ask HN: Alternatives to Reducto?
Looking for high-quality alternatives to Reducto for document parsing. Need strong table extraction + structured JSON output. Accuracy on messy PDFs is critical. What are people using in production?
Looking for high-quality alternatives to Reducto for document parsing. Need strong table extraction + structured JSON output. Accuracy on messy PDFs is critical. What are people using in production?
Comments
I've built a platform specialized for your use case: https://www.fabrx.ai/
You can setup your own custom extraction in under a minute without writing a single line of code. Free one month trial, no credit card required. Plus, the subscriptions are a flat rate, meaning you don't pay per page or number of documents processed.
Please try my platform: https://www.fabrx.ai/. I specially designed it for use cases like this. You can setup your custom extraction in under a minute, without writing a single line of code.
Try https://docstrange.nanonets.com/ once, 10k docs you can use for free. Strong table performance. Do give feedback if any. Powered by bigger model compared to our open source one which is quiet popular on HF.
Hey, worked on document parsing with tables (enterprise scale rag) at my prev job. What exactly are you trying to do? Can help with this
wanted to bump this. feel free to reply here or email me.
Hey I'm looking for high accuracy and high speed, its unclear to me if there is a good benchmark out there for this. Willing to try some open source solutions and self hosting as well.