Private Study Buddy for College Students with Sensitive Notes
A study and revision app that runs an LLM entirely on NPU-equipped student laptops, so students can summarize lectures, quiz themselves, and ask questions about their notes without their coursework being uploaded to cloud AI providers.
University students cramming for exams using lecture notes and PDFs
- Drop-in lecture PDFs and slides, get a quiz and flashcards generated locally
- 'Ask my notes' chat that runs entirely on-device
- Spaced-repetition scheduler built from your own material
- One-click 'study mode' that disables cloud network for the session
Students have been told to be careful about pasting course material into cloud AI; Ryzen AI Halo laptops give them an honest local option for the first time.
Multiple 2025 sources confirm real student privacy concerns about cloud AI (Stanford study, Axios, Springer paper), but willingness to switch behavior is unproven — most students keep pasting notes into ChatGPT anyway.Study exposes privacy risks of AI chatbot conversations ↗AI and ChatGPT use raises new fears for students' privacy - Axios ↗
AI study-app space is saturated — Quizlet, StudyFetch, Knowt, Laxu AI, NotebookLM, Monic.ai, Anki+third-party generators all compete; FastFlowLM and similar tools serve local-LLM devs but no consumer 'local-only study' app yet, leaving only a narrow niche.Laxu AI vs Quizlet vs Anki vs StudyFetch vs Knowt: Best Flashcard App in 2026? ↗6 Best AI Study Guide Generators in 2026 (Free & Paid) ↗
Clear WTP in category — StudyFetch ~$19/mo, Knowt Ultra $9.99/mo, Quizlet Plus $7.99/mo, Laxu $4.99/mo — but TAM is constrained to students who own Ryzen AI Halo (or future NPU) laptops, a small slice.Laxu AI vs Quizlet vs Anki vs StudyFetch vs Knowt: Best Flashcard App in 2026? ↗
Local LLM on NPU is a real, accelerating trend — AMD actively promoting hybrid NPU+iGPU LLM inference, Strix Halo getting serious hardware press, FastFlowLM shipping as a runtime, Apple/Qualcomm pushing same direction.Accelerate Fine-Tuned LLMs Locally on AMD Ryzen AI NPU & iGPU ↗Strix Halo, Unleashed: Real LLM Workflows on 128GB Ryzen AI ↗
AMD has demonstrated hybrid NPU+iGPU LLM inference and FastFlowLM provides a working runtime, but shipping a polished consumer study app that auto-selects models, handles PDFs/notes, and feels responsive on-device is still significant engineering; small-model quality for nuanced Q&A remains the risk.GitHub - FastFlowLM/FastFlowLM: Run LLMs on AMD Ryzen AI NPUs ↗AMD Targets Faster Local LLMs: Ryzen AI 300 Hybrid NPU+iGPU Approach ↗