Every product in this category claims the same things
Upload anything. AI flashcards. AI quizzes. AI notes. A tutor. The category has converged on an identical feature list, and the honest consequence is that a feature-by-feature table between any two of these products tells you almost nothing about which one will actually help you.
So compare the two properties that are hard to fake and that decide whether the output is worth anything.
Test one: ask it something your material doesn't cover
Upload a single lecture. Ask a question about a topic from a different week. Either it tells you the material doesn't cover it, or it produces a confident answer from the model's own knowledge — and if it does the second thing, you have just learned that you can never tell which mode any future answer was in. Do this with both products. It takes two minutes and it is more informative than any review.
Test two: check what 'spaced repetition' means to them
Ask what scheduling algorithm the flashcards use. "Spaced repetition" is claimed by nearly everyone and implemented properly by very few — most of it is a fixed 1-day/3-day/7-day ladder, which is a reminder schedule, not a memory model. Thaxas runs FSRS-4.5 with the published weights, tested against reference values. That is a checkable claim, and you should make every product in this category make one.
What we'll say about ourselves, plainly
| Thaxas | |
|---|---|
| Price | $12/mo, or $8/mo billed yearly. No free tier |
| Citations | Every claim cites the page in your upload |
| Refusal | Explicit 'not in your material', enforced at three layers |
| Scheduler | FSRS-4.5, published weights, tested against reference values |
| Grading | Short answers graded against a rubric, not string-matched |
| Privacy | Private storage, never used to train models, never shared |
| Mobile | Web app; no native iOS/Android app today |
_Written in July 2026 from publicly available information. These products ship constantly, so check their site before you decide — and tell us if we've got something wrong._