rember.com

Rember is an AI canvas that turns material you want to learn into spaced-repetition flashcards. You describe what you want to remember in natural language, and Rember generates cards built on practices from the spaced-repetition community. It fits learners who already study from many sources — books, videos, AI chats, pasted text — and want a faster path from raw material to reviewable cards. Cards can be studied inside Rember or reviewed in Anki. The site currently offers a waitlist and a direct call-booking option rather than open sign-up, so access is not immediate.

How Rember works

The workflow has three stages, as described on the site:

1. Learn everywhere. Read books, watch YouTube, chat with AI, or discuss with friends. Rember is designed to accept any input type.
2. Capture with the AI canvas. Tell Rember what you'd like to remember and give instructions in natural language. The canvas gives you control over what becomes a flashcard.
3. Review quality cards. Study the generated flashcards directly in Rember or in Anki.

The canvas accepts several input forms:

• Typed concepts, such as "linear algebra basics"
• References to books, such as a specific method from a named textbook
• Pasted text from anywhere
• URLs and YouTube videos
• AI chats, captured through an MCP integration

What "quality cards" means here

Rember states that flashcard quality is everything and that it implements practices developed by the spaced-repetition community over three decades. The site lists four principles:

• Test deep understanding rather than surface recall
• Make it atomic — one idea per card
• Make it unambiguous
• Avoid yes/no questions

The site illustrates the difference with a mitochondria example. A card asking "What's the function of mitochondria?" is weaker than one asking "Why are muscle cells packed with more mitochondria than other cell types?" — the second requires applying the concept rather than reciting a definition. This is the standard the generated cards aim for, and it is the main thing to check when you review output: whether cards actually probe understanding or just restate facts.

Reviewing and scheduling

Rember integrates FSRS, a spaced-repetition scheduling algorithm. Reviews can be paused, snoozed, or undone. The site says no setup is needed and that settings are configured for you, so you are not expected to tune scheduling parameters yourself.

If you are not ready to move your review workflow into Rember, an Anki add-on lets you generate cards in Rember and review them in Anki. There is also an MCP server that lets you say "add to Rember" inside Claude or ChatGPT chats.

Who it is for

Rember describes itself as built for serious learners — people who pull knowledge from PDFs, YouTube, AI conversations, and pasted text and want those sources converted into reviewable cards without hand-writing each one. It suits you if:

• You already use spaced repetition and care about card formulation
• Your study material is scattered across formats
• You want AI assistance but want to steer what becomes a card
• You are willing to wait for access or book a call

It is a poorer fit if you need an open, immediately usable tool today, or if you prefer to write every card by hand for full control.

What the site does not specify

The available site material does not state pricing, plan tiers, or whether a free tier exists. It also does not detail platform availability beyond the browser app, the Anki add-on, and the MCP integration. Treat cost and access terms as unknown until confirmed through the waitlist or a call.

Practical next step

Join the waitlist or book a call through the site. Before committing, prepare one source you actually study — a chapter, a lecture video, or a pasted article — and test whether the generated cards meet the four quality principles above. That single check tells you more about fit than any feature list.

Rember schedules reviews with the FSRS (Free Spaced Repetition Scheduler) algorithm, the open scheduling method the spaced repetition community has developed and refined. You don't configure it: Rember's settings are described as working out of the box, and your day-to-day control comes down to three actions — pause, snooze, or undo a review. That makes it a fit for learners who want scheduling handled for them rather than tuned by hand.

What FSRS does in Rember

FSRS decides when you see a card again based on your review history, rather than fixed intervals like "1 day, 3 days, 7 days." Rember's page states it integrates FSRS and that the product follows best practices from the spaced repetition community. In practice, this means:

• Your answers drive the schedule. Each review updates the model's estimate of how well you know that card, and the next due date follows from that estimate.
• You are not asked to pick intervals. There is no setup step described for choosing algorithm parameters.
• The schedule is per card, not per deck. Cards you find hard come back sooner; cards you know well stretch out.

The site does not publish the specific FSRS version, parameter defaults, or whether you can adjust desired retention. If exact scheduling control matters to you, treat that as an open question to confirm before committing.

The three controls you actually use

Rember's stated review controls are pause, snooze, and undo. These cover the common reasons people fall behind:

Undo is the one worth noticing. In most spaced repetition tools a wrong grade permanently shifts a card's interval; being able to reverse it means a slip doesn't distort your schedule.

Where the cards get reviewed

You can study generated cards directly in Rember, or export the workflow to Anki. Rember offers an Anki add-on for people who want to generate cards in Rember but keep reviewing in Anki. If you review in Anki, Anki's own scheduler governs your intervals — Rember's FSRS integration applies to reviewing inside Rember. Decide which tool holds your review history before you build a large deck, because that's where the scheduling lives.

How card quality affects scheduling

Scheduling only works if the cards are answerable. Rember says it builds cards around practices the community has settled on: test deep understanding, keep cards atomic, make them unambiguous, and avoid yes/no questions. The site's own example shows the difference:

• Weak: "What's the function of mitochondria?" → "Generating energy for the cell"
• Better: "Why are muscle cells packed with more mitochondria than other cell types?" → "Mitochondria generate energy, which muscles need for contraction"

The second card asks for a reason, so a correct answer is evidence you understand the mechanism. Vague or compound cards produce unreliable grades, and unreliable grades produce a schedule that doesn't match what you actually know.

Inputs that feed the review queue

Rember accepts typed concepts (for example "linear algebra basics"), references to specific books, pasted text, URLs and YouTube videos, and AI chats captured through an MCP integration — you can say "add to Rember" in Claude or ChatGPT. The AI canvas lets you direct in natural language what should become a card. More input types means more cards entering the same FSRS queue, so the review load grows with what you capture; pause and snooze are how you manage that load.

Is this the right fit for you?

Choose Rember's built-in scheduling if you want spaced repetition without tuning parameters and you're comfortable with pause/snooze/undo as your main levers. Choose the Anki route if you already have a review history or want Anki's scheduling controls — generate in Rember, review in Anki. Rember is currently invite-based: the site offers a waitlist or a direct call booking, so access isn't immediate. Pricing and plan limits are not stated on the page, so confirm cost and any account requirements before you rely on it for a semester or exam cycle.

Rember generates flashcards according to the spaced repetition community's established guidelines, and its site names four specific rules: test deep understanding, make each card atomic, keep wording unambiguous, and avoid yes/no questions. These apply to any card Rember produces from your input, whether you paste text, drop a PDF or YouTube link, type a concept, or capture an AI chat. If you want cards that probe reasoning rather than surface recall, these are the rules to check against.

The four rules Rember implements

Rember says it follows best practices "developed by the spaced repetition community over the past three decades." On its site, that translates into four concrete constraints:

The site frames this as a quality issue first: "Flashcard quality is everything." The rules exist because a card that can be answered by recognition or guessing produces weak review data, which undermines the scheduling that spaced repetition depends on.

The example Rember uses to show the difference

Rember's own contrast makes the "deep understanding" rule concrete. A surface-level card asks:

What's the function of mitochondria?

Answer: Generating energy for the cell.

That card can be passed by recalling a phrase. The deeper version asks:

Why are muscle cells packed with more mitochondria than other cell types?

Answer: Mitochondria generate energy, which muscles need for contraction.

The second card forces you to connect two facts — mitochondria make energy, muscles contract — and explain the relationship. That is the kind of card the first rule is meant to produce, and it also satisfies the other three: it is one idea, it has one clear answer, and it is not a yes/no question.

How the rules show up in the workflow

Rember's stated workflow is: learn anywhere (books, YouTube, AI chats, conversations), capture with the AI canvas by describing in natural language what you want to remember, then review the generated cards in Rember or in Anki. The best-practice rules are applied at the generation step, so the quality control happens before you ever see the card.

Two practical implications:

• You steer the depth. Because you give instructions in natural language, you can ask for reasoning-style cards ("why" and "how" prompts) rather than definition cards. The muscle-cell example shows what that looks like.
• You can review elsewhere. Rember says you can study generated cards directly in Rember or export the workflow to Anki, so the card-quality rules travel with the cards rather than being tied to one review app.

What to check before relying on it

The site describes the rules and gives one worked example, but it does not publish a full sample deck or detail how strictly each rule is enforced across every input type. If card quality is your main reason for choosing a tool, the useful test is to run one source you know well — a chapter or a lecture — through the canvas and inspect the output against the four rules above. Cards that fail the atomicity or ambiguity test are easy to spot, and you can then adjust your natural-language instructions to push for deeper prompts.

Rember also states it integrates FSRS for scheduling, with pause, snooze, and undo for reviews, and that no setup is needed. Those affect review behavior rather than card quality, but they matter if you plan to keep the whole loop inside Rember instead of Anki.

Yes. Rember connects to both Anki and AI chat tools, but in different ways. Its Anki add-on lets you generate cards in Rember and review them in Anki, so you don't have to leave Anki to benefit from Rember's card generation. Its MCP server and integration let you add material from Claude or ChatGPT by saying "add to Rember" in the chat. Both options are aimed at people who want Rember's AI card creation without fully switching tools.

How the Anki integration works

Rember's site describes the Anki add-on as an option for people who are "not ready for the jump": you generate cards in Rember, then review them in Anki. In practice, that means:

1. Input: Bring material into Rember through the AI canvas (typed concepts, pasted text, PDFs, URLs, YouTube videos, or captured AI chats).
2. Action: Tell Rember in natural language what you want to remember, and it generates flashcards following spaced-repetition best practices.
3. Output: Study the generated cards directly in Rember or in Anki.

The expected result is that your existing Anki review routine stays intact while card creation moves to Rember. The site does not detail add-on installation steps, so treat setup specifics as something to confirm inside Rember after signing in.

How the AI chat integration works

Rember offers an MCP server and integration for Claude and ChatGPT. According to the site, you "just say 'add to Rember'" in your chats. This fits a workflow where learning happens in conversation: you discuss a topic with an AI, then capture the useful parts as flashcard material without copying and pasting manually.

Captured AI chats are also listed as a supported input type in Rember's AI canvas, alongside PDFs, YouTube, and pasted text.

Which option fits you

What to keep in mind

• Card quality is the point. Rember says it implements spaced-repetition best practices: test deep understanding, make cards atomic, make them unambiguous, and avoid yes/no questions. Its example contrasts a vague "What's the function of mitochondria?" with a more specific question about why muscle cells have more mitochondria. The generated cards are meant to reflect this standard, but reviewing a sample before committing to a workflow is sensible.
• Review controls exist in Rember. The site mentions FSRS integration and the ability to pause, snooze, or undo reviews, plus that no setup is needed because settings are preconfigured.
• Access is waitlist-based. The site invites visitors to "Join the waitlist or book a call directly," so availability of the integrations may depend on getting access. Pricing and login requirements are not stated on the page, so don't assume either way.

If your main concern is keeping Anki, start with the add-on path. If your learning already happens in AI chats, the MCP route removes the most friction.

Rember turns your study material into flashcards through a three-step flow: bring in any input (typed concepts, books, pasted text, URLs, YouTube videos, or AI chats), tell the AI canvas in natural language what you want to remember, then review the generated cards in Rember or export them to Anki. The capture step is where you keep control — you decide which content becomes cards, and the generator follows spaced-repetition best practices so the output is atomic, unambiguous, and tests deep understanding rather than yes/no recall.

Step 1: Bring in whatever you're learning from

Rember accepts any input, so you don't need to reformat your material first. Supported sources per the site:

• Typed concepts — e.g., "linear algebra basics"
• Books — e.g., "sample-average method in Sutton Barto RL book"
• Pasted text from anywhere
• URLs and YouTube videos
• AI chats — via the MCP integration, you can say "add to Rember" inside Claude or ChatGPT

The idea is to capture at the moment you learn — reading, watching, chatting — rather than saving material for a later manual conversion pass.

Step 2: Direct the AI canvas in natural language

This is the core of the workflow. Instead of auto-dumping everything into cards, the AI canvas lets you specify what should become flashcards. You give instructions in plain language about what you'd like to remember, and Rember generates cards from that scoped request.

The site describes this as giving you "precise control over what becomes flashcards," which matters because indiscriminate card generation produces clutter you'll resent during review.

What "good" output looks like

Rember states it implements best practices from the spaced-repetition community built over three decades. The stated principles:

The site's own example shows the difference. A weak card asks "What's the function of mitochondria?" with the answer "Generating energy for the cell." A stronger card asks "Why are muscle cells packed with more mitochondria than other cell types?" — the answer connects structure to function ("Mitochondria generate energy, which muscles need for contraction"). When you write your capture instructions, phrasing them around mechanisms and comparisons tends to steer generation toward the second style.

Step 3: Review in Rember or Anki

Generated cards can be studied directly in Rember or sent to Anki. Rember integrates FSRS (a modern spaced-repetition scheduler), and you can pause, snooze, or undo reviews. The site notes no setup is needed for its settings.

If you already live in Anki, there's an Anki add-on path: generate cards in Rember, review them in Anki. That's the lower-commitment option if you don't want to switch review tools.

A concrete walkthrough

Say you're working through a reinforcement learning textbook and want to retain the sample-average method.

1. Capture: Reference the book by name ("sample-average method in Sutton Barto RL book") or paste the relevant passage.
2. Instruct: Tell the canvas what matters — e.g., that you want to remember how the method updates estimates incrementally and why it's used for stationary problems.
3. Generate and inspect: Check each card against the four principles above. Rewrite instructions if cards come out as yes/no or bundle multiple facts.
4. Review: Study in Rember with FSRS scheduling, or export to Anki.

Common sticking points

• Vague instructions produce vague cards. "Make flashcards from this chapter" gives the canvas little to work with; naming the specific concepts and the kind of understanding you want tested gives better results.
• Watch for non-atomic cards. If one card asks two things, split it — either by refining your instruction or editing after generation.
• Access note: the site currently offers a waitlist or a direct call booking, so availability isn't immediate. Plan around that if you're evaluating it for a course starting soon.

Is this the right fit?

Choose Rember if you want AI-assisted card creation with explicit control over what gets carded and plan to review in Rember or Anki. If you prefer building every card by hand, or you need a tool you can start using today without a waitlist, the capture-and-generate workflow won't save you the time you're looking for.