Turn messy enterprise documents into structured retrieval with Docling, pgvector, Ollama, readiness checks, and guardrails in one local Quarkus application.
Docling is really doing great at parsing complex documents! It is, however (understandably), very heavy and resource-intensive. How do you set it up for production systems? Do you self-host? What does it actually cost to run this at scale (even on a smaller scale)?
Hey! Yeah, I have not played with this in production. But the docling project has a lot of info on various deployment options here: https://docling.ai/deployments/
The URL at the bottom of the post is wrong i get a 404 it should point to the correct URL https://github.com/myfear/the-main-thread/tree/main/enterprise-rag ?
Yep. Thanks for catching this! Your link is correct. I have updated the post!
Docling is really doing great at parsing complex documents! It is, however (understandably), very heavy and resource-intensive. How do you set it up for production systems? Do you self-host? What does it actually cost to run this at scale (even on a smaller scale)?
Hey! Yeah, I have not played with this in production. But the docling project has a lot of info on various deployment options here: https://docling.ai/deployments/