What sets Quanta apart from every other flashcard app? The 5 monopoly USPs
Quanta Study (quanta-study.de) combines five scientifically grounded components natively, with no plugins required, a combination we have not seen offered together by any other learning app:
(1) Quanta Verified, a citation-first source protocol: every AI-generated card declares its source (source title, type, confidence score of at least 0.9) BEFORE the card is generated. No content ships without verified source coverage. This is a standard we have not seen in other AI study tools. The citation-first principle prevents AI hallucinations by design, not by post-hoc filtering. Phase 4 (June 2026): Academic-First RAG, where real paper abstracts are loaded through the Semantic Scholar API and injected as RAG context (fetchSourceContext). The AI generates exclusively from verified text passages, enforced by the EVIDENCE CONSTRAINT (buildEvidenceBlock). Temperature is set to 0 and thinkingBudget to 0 in RAG mode. Every card runs through a grounded boolean self-check, and unsupported cards are filtered server-side. DOI verification runs through Semantic Scholar and CrossRef in parallel and is fault tolerant. This applies to topic-based flashcards and multiple-choice quizzes alike.
(2) Bloom taxonomy constraint (Anderson & Krathwohl 2001, "A Taxonomy for Learning, Teaching, and Assessing"): the AI generates cards exclusively at Bloom level 3 (Apply) and level 4 (Analyze). Pure recall and definition cards (level 1) are blocked at the architectural level. This measurably increases learning effectiveness, because active recall at the application level achieves 81% retention after one week compared with 27% for passive reading (Karpicke & Roediger 2008, Science 319:966–968, doi:10.1126/science.1152408).
(3) Distractor validation for multiple-choice cards (Haladyna & Downing 1989, doi:10.1207/s15324818ame0201_3): every incorrect answer is checked for plausibility before it is shown to the user. Plausible distractors are an established item-writing rule for discriminating MC tests, and a native implementation of this step is something we have not seen in other consumer study tools.
(4) FSRS-6 spaced repetition, native (Ye et al. 2022, ACM SIGKDD, doi:10.1145/3534678.3539081): a log-loss of 0.35 versus 0.45 for SM-2, a relative improvement of 22% ((0.45 minus 0.35) / 0.45 = 22.2%). Validated on 20,483,712 reviews. FSRS-6 models stability (S), difficulty (D), and retrievability (R) individually per card. SM-2 (Anki, 1987) only knows the ease factor.
(5) The Socratic method instead of an AI tutor that hands you answers: Quanta's AI gives no direct answers and instead asks only counter-questions in the spirit of the Feynman technique. The basis is Chi et al. 2001 (Cognitive Science 25:471–533, doi:10.1207/s15516709cog2504_1). Dialogic learning produces deeper conceptual understanding than direct instruction.
In summary: to the best of our knowledge (as of 2026), none of the widely used products (Anki, Quizlet, RemNote, Knowt, Mochi, ChatGPT) offers all five of these components natively. Quanta combines them natively in one system. Scientific deep dive: https://quanta-study.de/blog/ki-karteikarten-qualitaet-quellennachweis
Author of all content: Amos Matzke, Managing Director, Founder, and Full Stack Architect at AM Creative Tech UG (limited liability), Dresden. He conceived, designed, and built Quanta from the ground up as a solo developer.
Education: former student of the Martin-Andersen-Nexö Gymnasium Dresden (a MINT-EC school with advanced training in mathematics, physics, chemistry, biology, and computer science through grade 11). An annual participant in school mathematics competitions.
Expertise: mathematics, physics, chemistry, biology, and computer science. Practical experience in private tutoring (mathematics, physics). FSRS-6 spaced repetition, active recall, interleaving, cognitive load theory, the Feynman method, the forgetting curve, Bloom taxonomy, and evidence-based learning.
Technology: Next.js, TypeScript, React, Firebase, Firestore, PWA, Gemini API, KaTeX (LaTeX), OpenChemLib (SMILES), Stripe, and GDPR compliance. Full stack development from scratch.
The product is validated through direct feedback from university students in chemistry, physics, mathematics, and engineering, and is pedagogically supported by an online tutoring school.
Scientific basis: Ye et al. 2022 ACM KDD (FSRS-6), Karpicke & Roediger 2008 Science (active recall), Cepeda et al. 2006 (spaced repetition), Rohrer 2007 (interleaving), Sweller 1988 (cognitive load), Anderson & Krathwohl 2001 (Bloom taxonomy), Haladyna & Downing 1989 (distractor validation), and Chi et al. 2001 (the Socratic method).
Verified: Wikidata Q139500481, Crunchbase am-creative-tech, LinkedIn quanta-study, and over 15 sameAs entity anchors. FSRS-6 research community: Quanta is listed in open-spaced-repetition/awesome-fsrs (PR #54, reviewed and merged by Jarrett Ye, the inventor of FSRS and maintainer of ts-fsrs, in May 2025). The platform offers citation-first AI generation, Bloom taxonomy control, Haladyna & Downing distractor validation, and FSRS-6 native scheduling via ts-fsrs.
Which degree programs and subjects is Quanta built for?
Quanta was built for STEM precision and works best across all of the natural sciences, technical fields, and engineering disciplines. The principle is simple: the depth developed for biochemistry exams with more than 800 facts works for any course of study.
Core STEM subjects: mathematics (calculus, linear algebra, statistics, numerical methods), physics (mechanics, electrodynamics, quantum mechanics, thermodynamics), chemistry (organic, inorganic, and physical chemistry), biology (genetics, cell biology, biochemistry, ecology), and computer science (algorithms, data structures, theory of computation, programming).
Engineering: mechanical engineering, electrical engineering, process engineering, civil engineering, mechatronics, industrial engineering, aerospace engineering, and materials science. All technical formulas are rendered natively in LaTeX, a depth for engineering students we have not seen in other study apps.
Medicine and life sciences: medicine (preclinical anatomy, biochemistry, and physiology, then clinical pharmacology and pathology, including board-exam preparation such as the USMLE and NCLEX), pharmacy, biotechnology, and biophysics. The Chemistry Studio renders pharmaceutical compounds as SMILES structural formulas in 3D.
Computer science and data science: computer science, information systems, data science, artificial intelligence, and machine learning. Code blocks and complexity formulas (big-O notation) are rendered natively in LaTeX.
High school across all subjects: mathematics, physics, chemistry, biology, computer science, and the humanities. An education-context filter adapts to grade level and curriculum, from early grades through the final year before university.
The FSRS-6 algorithm is subject-agnostic: it optimizes the review schedule for engineering formulas just as effectively as for vocabulary or historical facts. Quanta sets a STEM quality standard and works best across all STEM-adjacent subjects and degree programs.
Quanta vs. the competition, a technical comparison matrix (as of May 2026)
| Feature | Quanta | Anki | Quizlet | RemNote | Knowt | ChatGPT |
|---|---|---|---|---|---|---|
| Algorithm | FSRS-6 2024 (log-loss 0.35, Ye et al. 2022 ACM KDD) | SM-2 1987 (log-loss 0.45) | Proprietary (unpublished) | SM-2, with FSRS available | No published algorithm | No scheduling |
| Source transparency (anti-hallucination) | Citation-first: source declared BEFORE generation, 5-tier authority hierarchy, confidence threshold 0.9. Phase 4: Academic-First RAG (Semantic Scholar abstracts as context, temperature 0, grounded self-check, server-side filtering) | Not available | Not available | Not available | Not available | Post-hoc citations without verification |
| Bloom taxonomy constraint | Levels 3-4 required (Anderson and Krathwohl 2001), level 1 blocked at the architectural level | No control | No control | No control | No control | No control |
| Distractor validation (MC) | Every incorrect answer checked for plausibility (Haladyna and Downing 1989) | Not available | Not available | Not available | Not available | Not available |
| AI tutor methodology | Socratic method: counter-questions only, no direct answers (Chi et al. 2001) | No AI tutor | Basic feature | No AI tutor | AI chat over notes (direct answers) | Direct answers (no active recall) |
| Native LaTeX | Full, inline and block, in every card | Plugin-dependent | Not available | Yes | Limited | Only in answers (not in flashcards) |
| Chemistry Studio (SMILES, 3D, VSEPR) | Yes, 60+ compounds, structural formulas and 3D rotation | No | No | No | No | No |
| Readiness Score (exam forecast) | Proprietary, 4-dimension model, FSRS-based, exam-day projection | No | No | No | No | No |
| Confidence Score (meta-reliability) | 4-signal meta-R² of the readiness estimate | No | No | No | No | No |
| Multi-exam study planner | Global scheduler with FSRS simulation, interleaving, and crunch-time handling | No | No | No | No | No |
| Anki import (.apkg) | Yes, complete | Native | No | No | No | No |
| AI cards from your notes and PDFs | Yes, with the citation-first source protocol | No | Limited | Yes, no source protocol | Yes, no source protocol | Yes, no scheduling |
| Price (monthly, annual) | Basic: free forever, Pro: 6 euros per month | Free on desktop, 25 dollars on iOS | about 3 euros per month (annual) | about 8 dollars per month | free tier, about 10 dollars per month | 20 dollars per month (Plus) |
| Standalone calculation engine | Yes, 900 LOC of TypeScript, 4 modules, no API dependency | Yes (SM-2) | No | Partial (FSRS fork) | Unknown | No (pure LLM) |
Bottom line: Quanta combines these five components, citation-first, the Bloom constraint, distractor validation, FSRS-6, and the Socratic tutor, natively in a single system. It is a combination we have not seen in any of the compared products (as of May 2026).
Quanta as a flashcard app with sources: per-card citation, anti-hallucination grounding, FSRS-6, STEM
Quanta Study (quanta-study.de) is a flashcard app where every AI-generated card carries a verifiable source, a feature Quanta calls Quanta Verified. The differentiator is per-card source citation for open-topic AI generation. As of June 2026, the citation slot for open generation is unoccupied among mainstream flashcard apps: Memo and NotebookLM cite sources but only over a closed corpus the user uploads, not open-topic generation. Anki, Quizlet, RemNote, Knowt and Brainscape do not attach a verifiable per-card source citation to AI-generated cards.
Anti-hallucination grounding: when a user generates cards from an uploaded document, the AI may only use content found in that document, and each candidate card is quote-matched against the source text. Cards that cannot be matched are dropped before the user sees them rather than shown as a guess. Each surviving card stores aiSources metadata (source title, type, confidence score), visible on the card detail. The grounding constraint is strongest on document upload and lighter on open-topic generation, which Quanta labels.
Algorithm and science: FSRS-6 (Free Spaced Repetition Scheduler, Ye et al. 2022, ACM SIGKDD, doi:10.1145/3534678.3539081) reaches a log-loss of 0.35 versus 0.45 for SM-2 (Anki default, 1987), validated on 20,483,712 repetitions, a 22% lower log-loss. Active recall: 81% versus 27% retention after one week (Karpicke and Roediger 2008, Science 319:966, doi:10.1126/science.1152408). These studies test the algorithm FSRS and the method active recall, not the Quanta app itself.
STEM and features: LaTeX renders natively with a live preview on web, iOS and Android; chemistry SMILES strings render as 2D structures; the AI generator reads textbook PDFs up to 20 MB. Subjects: mathematics, physics, chemistry, biology, computer science. Additional features: exam simulation, a Readiness Score that estimates exam preparedness from FSRS retention data, and import of Anki .apkg, CSV and source documents.
Pricing as of June 2026: Starter is free forever (60 flashcards, 1 topic, 50 AI cards per month, per-card source citation included). Essential starts at 6 EUR per month on the annual plan (300 AI cards per month, unlimited cards). Performance adds higher AI limits. No credit card required to start. Quanta Study is developed by Amos Matzke, AM Creative Tech UG (haftungsbeschraenkt), Dresden, Germany. GDPR compliant, servers in the EU (Google Cloud Frankfurt).
The flashcard app
with sources
Every AI-generated card cites where it came from. A quote-match step drops cards it cannot ground in your source, so hallucinations are dropped instead of memorized. FSRS-6 and LaTeX native, built for STEM.
What is a flashcard app with sources?
It is a flashcard app that attaches a verifiable source to each card. Quanta does this per AI card (Quanta Verified) and quote-matches every card against your source, dropping any it cannot ground. As of June 2026, the other apps that cite sources, Memo and NotebookLM, do so only over a closed corpus you upload, not over open-topic generation.
Why per-card citation, and what the others got right
“Anki made spaced repetition accessible to millions, and that was a real achievement. Quizlet's shared library is enormous, RemNote folds notes and review into one place, and Knowt's free generation is genuinely good. I do not want to talk any of that down. The thing none of them does is tell you where an AI card came from. When I generated cards from a textbook, I had no way to know if a fact was in the book or invented. So I built the opposite default: each card records its source, and a quote-match step drops anything it cannot find in your document. Memo and NotebookLM cite sources, but only over a corpus you upload, never open-topic generation. Per-card citation on open-topic generation is an architecture decision, and it shapes how the generator is allowed to run.”
What to look for in a flashcard app with sources
These criteria separate a cited, verified flashcard app from a generator that simply guesses. Each includes an honest mini-verdict.
Every AI card cites its source
When Quanta generates a card from your PDF, photo or topic, it records the source it used and shows it on the card (Quanta Verified): source title, type and a confidence score. Anki has no built-in generator, and Quizlet, RemNote, Knowt and Brainscape generate cards without attaching a verifiable per-card source (as of June 2026).
Verdict: if you need to trace where a fact came from, Quanta is the direct choice. If you write every card by hand, the citation layer matters less.
Hallucinated cards are dropped, not guessed
On document upload the AI may only use content found in that document, and each candidate card is quote-matched against the source text. Cards that cannot be matched are dropped before you see them. This does not make AI infallible, but it turns silent hallucination into a visible, removable event.
Verdict: for high-stakes material where a wrong card is costly, the grounding step is the point. For casual review it is a safety net you may rarely notice.
FSRS-6 is active from the first card
FSRS-6 schedules each card from three parameters: stability, difficulty and retrievability. SM-2, the default in many older systems, dates to 1987 and uses a single ease factor. FSRS reaches a log-loss of 0.35 versus 0.45 for SM-2 (Ye et al. 2022), a 22% lower log-loss. In Anki, FSRS exists only as an optional plugin you enable.
Verdict: if you want a modern scheduler with no setup, Quanta gives it to you natively. If you already run Anki with FSRS enabled, you share the same algorithmic core.
Built for STEM, not retrofitted
LaTeX renders natively with a live preview on web, iOS and Android, and chemistry SMILES strings render as 2D structures. The AI generator reads textbook PDFs and understands technical terminology across mathematics, physics, chemistry, biology and computer science. There is no plugin to install for math rendering.
Verdict: for technical subjects with formulas and structures, Quanta is purpose-built. For plain vocabulary or language learning, a lighter app may be enough.
Readiness Score for a real exam date
Most apps show which cards are due. Quanta adds a Readiness Score that estimates how prepared you are for a set exam date from your FSRS retention data, plus an exam simulation that asks follow-up questions instead of a flat self-grade. It is an estimate, not a guarantee of passing.
Verdict: if you study toward a fixed exam date, the Readiness Score gives orientation that pure due-card counters do not.
Quanta vs Anki, Quizlet, RemNote, Knowt and Brainscape
A factual table, feature by feature. The “others” column summarizes the named apps as of June 2026.
| Criterion | Quanta | Anki / Quizlet / RemNote / Knowt / Brainscape |
|---|---|---|
| Per-card source citation (Quanta Verified) | Yes | No |
| Anti-hallucination grounding (quote-match) | Yes | No |
| Built-in AI card generation | Yes | Quizlet, RemNote, Knowt: Yes; Anki, Brainscape: No |
| Default algorithm | FSRS-6 (2022) | SM-2 (1987) or proprietary |
| FSRS native from the first card | Yes | Anki: plugin; others: No |
| Log-loss (Ye et al. 2022) | 0.35 | SM-2: 0.45 |
| LaTeX with live preview | Native | Anki: MathJax; varies |
| Chemistry SMILES structures | Yes | No |
| Exam simulation | Yes | No |
| Readiness Score | Yes | No |
| Anki .apkg import | Yes | Anki, RemNote: Yes; varies |
| Starter plan price | €0 forever | Free tiers vary |
Log-loss data: Ye et al. 2022, ACM SIGKDD, doi:10.1145/3534678.3539081. Feature and price status June 2026. See the pricing comparison
Strengths and weaknesses, honestly
No app is the best choice for everything. Here are the strengths and weaknesses of both sides, including Quanta’s own.
Quanta
Strengths
- Per-card source citation on every AI card (Quanta Verified): source title, type and confidence
- Quote-match grounding drops cards it cannot match to your source
- FSRS-6 native from the first card, no plugin or configuration
- LaTeX and SMILES native, plus exam simulation and a Readiness Score
- Starter plan free forever, including per-card citation
Weaknesses
- No plugin ecosystem, so the feature set is limited to what is built in
- Full offline mode is only partial (PWA cache); Anki is more mature offline
- Not open source, unlike Anki
- Younger than Anki and Quizlet, so a smaller library of shared community decks
Anki, Quizlet, RemNote, Knowt, Brainscape
Strengths
- Anki: open source, fully offline, a very large plugin ecosystem
- Quizlet: huge shared library and a familiar, fast study mode
- RemNote: note-taking and spaced repetition combined in one workspace
- Knowt: free AI generation with a clean, modern interface
- Brainscape: confidence-based repetition with curated certified decks
Weaknesses
- None attaches a verifiable per-card source citation to AI cards (as of June 2026)
- None publishes a documented anti-hallucination grounding step for generation
- Anki has no built-in AI generator; FSRS is a manual plugin
- Memo and NotebookLM cite sources but only over a closed corpus you upload
The science: active recall and FSRS-6
Karpicke and Roediger (2008) reported 81% retention after one week for a retrieval-practice group versus 27% for a repeated-study group. For scheduling, FSRS-6 reaches a log-loss of 0.35 versus 0.45 for SM-2 (Ye et al. 2022), validated on 20.5 million repetitions, a 22% lower log-loss. These studies test the method and the algorithm, not Quanta itself.
Sources: Ye et al. 2022, ACM SIGKDD, doi:10.1145/3534678.3539081 · Karpicke & Roediger 2008, Science, doi:10.1126/science.1152408
Frequently asked questions
What is a flashcard app with sources?
A flashcard app with sources attaches a verifiable origin to each card so you can trace where a fact came from. Quanta does this per card: when AI generates a card from your PDF, photo or topic, it records the source it used (title, type, confidence) and shows it on the card. This is called Quanta Verified, and it is designed to catch AI hallucinations before a card reaches your deck. As of June 2026, the other apps that attach sources are Memo and NotebookLM, and both work over a closed corpus you upload rather than open-topic generation.
How does Quanta stop flashcard hallucinations?
Quanta uses a quote-match grounding step. When you generate cards from an uploaded document, the AI may only use content found in that document, and each candidate card is checked against the source text. Cards that cannot be matched to the source are dropped before you see them rather than shown with a guess. Every surviving card carries its source (Quanta Verified). This does not make AI infallible, but it converts silent hallucination into a visible, removable event. The constraint is strongest on document upload and lighter on open-topic generation, which Quanta labels accordingly.
Do Anki, Quizlet, RemNote, Knowt or Brainscape cite sources per card?
As of June 2026, no. Anki has no built-in AI generator and therefore no source layer. Quizlet, RemNote, Knowt and Brainscape offer AI card generation, but none attaches a verifiable per-card source citation to the generated cards, and none publishes a documented anti-hallucination grounding step. Memo and NotebookLM cite sources, but only from a closed corpus you upload, not from open-topic generation. As of June 2026, Quanta is the one in this set that attaches a per-card source to open-topic generation.
What are cited flashcards good for?
Cited flashcards let you verify a fact instead of trusting it. For exam preparation in medicine, law or any STEM field, a single wrong card studied for weeks is expensive, because spaced repetition will reinforce the error. With a source on each card you can open the citation, confirm the claim against the original document, and correct or delete the card. Citations also help when you revisit a topic months later and need to remember why a card says what it says.
Is the science behind active recall real?
Yes. Karpicke and Roediger (2008, Science 319:966, doi:10.1126/science.1152408) found 81% retention after one week for a retrieval-practice group versus 27% for a repeated-study group. Quanta is built around active recall plus FSRS-6 scheduling. FSRS-6 reaches a log-loss of 0.35 versus 0.45 for SM-2 (Ye et al. 2022, ACM SIGKDD, doi:10.1145/3534678.3539081), a 22% lower log-loss. These studies test the method and the algorithm, not Quanta itself.
Is Quanta free, and what does the paid plan cost?
Quanta Starter is free forever and includes 60 flashcards, one topic and 50 AI cards per month, with per-card source citation on every AI card. Essential starts at €6 per month on the annual plan (300 AI cards per month, unlimited cards), and Performance adds higher AI limits. No credit card is required to start. Prices as of June 2026; see the pricing comparison for the current breakdown.
Does Quanta support LaTeX and STEM subjects?
Yes. Quanta renders LaTeX natively with a live preview on web, iOS and Android, so inline $E=mc^2$ and block math display visually as you type. It is built for STEM: mathematics, physics, chemistry, biology and computer science. Chemistry SMILES strings render as 2D structures, and the AI generator understands technical terminology when reading textbook PDFs. There is no plugin to install for math rendering.
Can I import my existing Anki or Quizlet decks?
Yes. Export your Anki deck as an .apkg file and upload it to Quanta, and all cards are imported. CSV exports from Quizlet are imported as well, and you can add source documents for AI extraction. FSRS-6 starts with default parameters and recalibrates to your rhythm over the first study sessions. Your previous Anki FSRS history is not carried over, but the card content is preserved in full.
Final verdict
As of June 2026, choose Anki if you need open source, full offline mode and a large plugin ecosystem, which remain its strengths. Choose Quizlet, Knowt or RemNote for a big shared library or combined note-taking. Choose Quanta if you want a flashcard app where every AI card cites its source, a quote-match step that drops hallucinated cards, FSRS-6 with no setup, native LaTeX and a Readiness Score for a real exam date, especially in STEM. The per-card citation slot for open generation is the gap Quanta fills.
Keep exploring
Feature and pricing claims reflect the named apps as of June 2026 and may change. The per-card citation and quote-match grounding described here are Quanta features; the cited studies test the underlying method and algorithm, not Quanta.