Wanted to share this one because I find myself using it every day—often in place of Google. You can even somewhat replace Siri with it using the app’s Apple Shortcuts:
Note: I haven’t read the privacy policy or terms and do suspect that there is some form of data collection.
Yeah that would be awesome, I couldn’t get this one to search scihub but the Academic mode does search more reputable sources
I’m building a custom search AI as well, but it would be pretty big to host a scihub search LLM.
It says there are 88,343,822 articles. For an AI to work effectively, you’ll have to slice up the articles into paragraphs, so you will probably end up with between 10x to 100x slices. For those slices you’d have to get the embed vector and store it in a Vector database.
One 1536 vector is about 6.15 KB, meaning 54331450530 KB for everything, or 543 GB in vectors