Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Customizable, embeddable Chat GPT based on your own documents

libraria.dev

Post reply on HN
Print view View original

81–90 of 135 posts

Re: Show HN: Customizable, embeddable Chat GPT based on your own documents

#81

Post by highwaylights »

On the website it states:

“Step 1

Import or sync documents into Libraria, or add API integrations like Google and Shopify (beta). Bring the docs - let GPT-3.5 do the heavy lifting.”

I might be reading it wrong, or might have missed it on the website, but is it actually GPT-3.5 running over those imported documents? (As in, are you using OpenAI or another third party provider in the background?).

If you’re running a local LLM then the privacy implications are clearly pretty different than if you’re essentially sending people’s documents verbatim into an external LLM.

Re: Show HN: Customizable, embeddable Chat GPT based on your own documents

#83

Post by bealuga »

dash2 wrote: post #44

I wonder if you’ve considered academics as a target market. We have a lot of pdfs and might like help in “thinking” about them.

That would be really really cool if I could able to serve that space. I'd be curious to know what kind of features you'd want to have, what would be deal-breakers, etc!

Re: Show HN: Customizable, embeddable Chat GPT based on your own documents

#85

Post by calculated »

bealuga wrote: post #72
calculated wrote: post #68

Great service, Bea! Would it support Databases as a source in the future?

Yes! I'm currently working on it. I'm trying to improve CSV imports right now (current users have large csvs they want to go through)

Amazing. Another thing - make it si that we can give you our emails and you can send us updates there so we can track the progress of the app and come back to it when needed.

Re: Show HN: Customizable, embeddable Chat GPT based on your own documents

#86

Post by highwaylights »

mat_jack1 wrote: post #66
intalentive wrote: post #56

I did this for my email and Google docs. It’s about 15 lines of Python with langchain.

Is this Ok to upload all your messages and your contacts to these services? I'm very worried about that, really

I’d assume it has exactly the consequences you think it does, but I don’t know that and it becomes more of an unknown as more third parties are layered between you and the LLM.

I’d assume the right prompt by another user of the same underlying LLM on another platform could well expose your private information / content / passwords and would treat this with a lot of caution until you get satisfactory evidence otherwise.

Re: Show HN: Customizable, embeddable Chat GPT based on your own documents

#87

Post by bealuga »

calculated wrote: post #85
bealuga wrote: post #72

Earlier quoted context omitted.

Yes! I'm currently working on it. I'm trying to improve CSV imports right now (current users have large csvs they want to go through)

Amazing. Another thing - make it si that we can give you our emails and you can send us updates there so we can track the progress of the app and come back to it when needed.

It's at the very bottom of the page! I'll try to make it more visible tomorrow as well (:

Re: Show HN: Customizable, embeddable Chat GPT based on your own documents

#89

Post by monkeydust »

highwaylights wrote: post #81

On the website it states: “Step 1 Import or sync documents into Libraria, or add API integrations like Google and Shopify (beta). Bring the docs - let GPT-3.5 do the heavy lifting.” I might be reading it wrong, or might have missed it on the website, but is it actually GPT-3.5 running over those imported documents? (As in, are you using OpenAI or another third party provider in the background?). If you’re running a l…

Also curious before I use.

Tbh I have managed to build this minus the nicer interface using langchain. Was surprisingly easy as someone who doesn't dev daily.

https://blog.langchain.dev/retrieval/

Re: Show HN: Customizable, embeddable Chat GPT based on your own documents

#90

Post by MaxikCZ »

bealuga wrote: post #83
dash2 wrote: post #44

I wonder if you’ve considered academics as a target market. We have a lot of pdfs and might like help in “thinking” about them.

That would be really really cool if I could able to serve that space. I'd be curious to know what kind of features you'd want to have, what would be deal-breakers, etc!

Not person with original question but he asks for the same feature as I would like to see.

I am not sure how the documents are handled in your product, since Chat GPT has a context limit that probably wont be able to hold longer papers in memory.

For me, I have a pdf[0] depicting a system that can be programmed, along with bits of pseudo-code and a lot of clarifications. Something that the Chat GPT could use to spit out an actual implementation, if it were able to "think" about the pdf as a whole. I would love to see if your product is capable of such feat.

[0]: https://arrow.tudublin.ie/cgi/viewcontent.cgi?article=1177&c...

Drive-Based Utility-Maximizing Computer Game Non-Player Characters by Colm Sloan (note that basically only chapter 3 is needed in this case, but its still over 40 pages long)

Post reply on HN