Deck options and presets

Repeto separates a reusable study preset from the private copy used by an individual deck. This makes global templates convenient without letting a later preset edit silently alter an established deck.

Global presets

Open Settings → Study options to create, clone, rename, delete, or edit presets. The seeded Default preset is the source for new decks and cannot be renamed or deleted.

A preset contains the complete scheduling template:

• new cards and maximum reviews per day;
• desired retention;
• learning and relearning steps;
• maximum interval;
• leech threshold and action;
• new/review ordering;
• automatic burial of eligible new, review, and interday-learning siblings;
• whether new cards ignore the review limit;
• custom FSRS-6 weights. Anki-style review and learning fuzz is always applied automatically, so there is no deck toggle that can accidentally disable it.

The default maximum review limit is 9,999, which is effectively “show all due reviews” for most collections. It is a Repeto product default, not a claim that 9,999 is Anki’s universal preset value.

Every deck gets a copy

When a deck is created, Repeto records its source preset and stores a complete private scheduling copy. Open the deck’s action menu and choose Settings to edit that copy on a full page.

The page states You are editing this deck’s copy. Saving affects only the deck (or, if explicitly selected, its subdecks). It does not edit the named global preset.

This design supports two common workflows:

• use one preset as a starting point for several subjects, then tune each deck;
• keep stable deck behavior even when the reusable preset evolves later.

Sync from preset

If the source preset changes after the deck copy was made, the deck shows that updates are available. Choose Sync from preset, then confirm Replace the copy.

Sync is a full replacement of every scheduling field in the deck copy. It is not a field-by-field merge. Today-only limits remain separate. If the deck has important local tuning, record it before replacing the copy.

Choosing another preset loads that preset as a new starting point. Save the deck copy to make the choice effective.

Apply to subdecks

On a parent deck, enable Apply to subdecks before saving or syncing to copy the selected source and every scheduling value to visible descendants. This is intentionally explicit because descendants may have their own learning steps or retention targets.

Today-only limits

The deck settings page lets you temporarily replace New cards per day and Maximum reviews per day. These values expire at the next scheduling rollover. Clear both fields to use the stored deck copy again.

Custom study → Increase today’s limits is a different operation. It additively raises the ordinary deck’s new or review capacity, caps each category at 9,999, expires at rollover, and can be reset early. It does not replace the stored deck copy and does not create a filtered deck.

Open Settings → Scheduling day to choose the account’s IANA timezone and a rollover hour from 0 through 23. Repeto uses that DST-aware scheduler day for daily limits, today-only overrides, burial expiry, set-due actions, and study statistics. Rollover precision is currently one whole hour.

Learn ahead

Open Settings → Study to set the account-wide Learn ahead limit from 0 to 1,440 minutes. The default is 20 minutes. A value of 0 waits for the full learning delay; a positive value lets a future intraday learning card appear early only after ordinary due learning, review, and new work is exhausted. Learn-ahead cards do not consume the requested session-card limit.

Hierarchical limits

A parent-deck session respects the daily new and review quotas of each concrete descendant. When Limits start from top is enabled, directly studying a child also applies the available quota of its concrete ancestors. Missing virtual ancestors in a :: path do not invent a quota, while an existing parent does.

Choosing values

Start with the Default copy and change one variable at a time. In particular:

• increase new cards only when you can sustain the later review load;
• increase desired retention only if the extra repetitions are worth it;
• keep learning steps in whole minutes, with no more than ten steps;
• use custom FSRS weights only when they came from a trustworthy optimization workflow;
• set a leech action so repeatedly failed material becomes visible for repair.

For the exact queue behavior and remaining option gaps, see Studying with Repeto and Anki parity roadmap.