News for hackers. Stuff that matters.

A comment on Show HN: Documind – Open-source AI tool to turn documents into structured data.

Re:Show HN: Documind – Open-source AI tool to turn documents... by inexcf on Monday November 18, 2024 @07:51AM (#42171899)
Got excited about an open-source tool doing this.

Alas, i am let down. It is an open-source tool creating the prompt for the OpenAI API and i can't go and send customer data to them.

I'm aware of https://github.com/clovaai/donut so i hoped this would be more like that.

[ Parent | on HN ]
Re:Show HN: Documind – Open-source AI tool to turn documents... by _joel on Monday November 18, 2024 @07:59AM (#42171944)
You can self host OpenAPI compatible models with lmstudio and the like. I've used it with https://anythingllm.com/
[ Parent | on HN ]
Re:Show HN: Documind – Open-source AI tool to turn documents... by turblety on Monday November 18, 2024 @08:03AM (#42171963)
You might be able to use Ollama, which has a OpenAI compatible API.
[ Parent | on HN ]
Re:Show HN: Documind – Open-source AI tool to turn documents... by Zambyte on Monday November 18, 2024 @08:17AM (#42172054)
Not without chaning the code (should be easy though)

https://github.com/DocumindHQ/documind/blob/d91121739df03867...

[ Parent | on HN ]
Re:Show HN: Documind – Open-source AI tool to turn documents... by Tammilore on Monday November 18, 2024 @08:39AM (#42172184)
Hi. I totally get the concern about sending data to OpenAI. Right now, Documind uses OpenAI's API just so people could quickly get started and see what it is like, but I’m open to adding options and contributions that would be better for privacy.
[ Parent | on HN ]
Re:Show HN: Documind – Open-source AI tool to turn documents... by inexcf on Tuesday November 19, 2024 @05:01PM (#42188570)
That sounds great.
[ Parent | on HN ]
Re:Show HN: Documind – Open-source AI tool to turn documents... by sidmo on Wednesday November 20, 2024 @12:00PM (#42195901)
I'd recommend checking out vision language models. They generate embeddings of the images themselves (as a collection of patches) and you can see query matching displayed as a heatmap over the document. Picks up text that OCR misses. I built a simple API over it if you want to try it out: https://github.com/DataFog/vlm-api
[ Parent | on HN ]