Can OpenAI Help Power a Public-Interest Library AI Project?

Can’t hurt to ask, right?

I’m a furloughed federal employee and I’ve been trying to put my time-off to good use. I’ve been vibe-coding a project to enable better discovery of a collection of over 150,000 public domain texts at the Library of Congress. I completed a version 1 (or maybe version .1) yesterday. I’m using batch-processing with GPT-5 mini for the text analysis, but even at those cheap costs, I can only afford to run analysis for 1000 books. I’d love to do the other 150,000. Is there anyone I could contact at OpenAI for possible assistance? Here’s the site so far (yes, it’s a little rough around the edges): https://life-explorer.replit.app/

If anyone wants to provide feedback, I’d love to hear it.

2 Likes

Greetings.

Just in case you haven’t heard of this, you can see if you are eligible for complimentary tokens by sharing data with openai. I will allow you to run a daily limited quota for free, which might be of some help in your project.

You can find the full details here.

I did not know this. Thanks.

I couldn’t figure out why my API costs always seemed to be lower than what I spent. I was already set up to get the free tokens. Nice. Once I’ve spent a bit more I’ll get a lot more free tokens. Kind of makes it worth while to spend some more API money. Thanks again!

1 Like