I have three massive 500-page textbooks for my comparative politics class and I'm honestly drowning in the reading list. My midterms are coming up in like ten days and there is just no physical way for me to get through all these chapters while also working my part-time job at the library. I've been trying to find a solid AI tool that can actually handle long academic PDFs without hitting a paywall five minutes in but its been a struggle. Most of the stuff I find online has these tiny file limits or they want a monthly subscription that I just cant afford right now.

I am currently stuck between a few options and I'm trying to figure out which one is the least likely to hallucinate or cut me off mid-sentence:

• ChatPDF: I tried it with a smaller article and it was okay but I'm worried it's gonna lose the plot when I upload a 60MB textbook. Does it actually index the whole thing or just the first few pages?
• Claude: Everyone keeps talking about the context window being huge but the free tier limits on messages are kind of annoying when you're trying to dig deep into a chapter.
• NotebookLM: I just heard about this one from a friend but I havent used it yet. It’s supposed to be good for sources but I dont know if it handles dense academic jargon well.

My main constraints are:

• Budget: $0. Total. I'm a grad student in Seattle and my budget for extra tech is basically non-existent.
• Accuracy: This is the big one. If the AI makes up a theory or misattributes a quote I am going to fail this exam.
• Depth: I dont just need a 3-sentence summary. I need it to actually explain the core arguments and the evidence used in each chapter.

I’m leaning towards NotebookLM because it’s Google and seems free for now but I’m worried it might be too beta and glitchy. Has anyone tried running full textbooks through these? Or is there a better workflow like maybe splitting the PDFs into chapters and using Claude? Just really need a way to get the main points without the fluff...