Minecraft Enchantment Order Calculator

It answers the question you actually have in front of the anvil: I own this sword and these four books — what goes in first?

Not "what does Sharpness cost". You enter the gear, anything already on it, how many times it has been through an anvil, and then each enchanted book you hold. What comes back is a numbered plan: step one, step two, step three, what each one charges in experience levels, what the whole thing costs, and which single operation is the most expensive.

It also tells you the thing a cost table cannot. If any step would go over the anvil's limit it says which step, and it looks for a different order that avoids it — even when that order costs more overall. If no order avoids it, it says so instead of showing you a total you could never pay.

It also refuses one thing on purpose. If two of your books can never end up on the same item — a Sharpness book and a Smite book, say — it will not quietly pick one and call the other unnecessary. Those are two different swords, and which one you want is your decision rather than an arithmetic one.

You can keep a handful of gear plans in this browser, so the set of books you are slowly collecting for a helmet is still there next week. Nothing is sent anywhere and there is no account.

Every anvil operation charges three things. The prior work penalty on the left-hand item, the prior work penalty on the right-hand one, and the enchantments that transfer. This page shows all three separately for every step, because they are what you change when you change the order.

The prior work penalty doubles. Every time something goes through an anvil, its use count goes up by one, and the penalty that count carries is 2 to the power of it, minus one: 0, 1, 3, 7, 15, 31. The result of an operation inherits the higher of the two counts, plus one. So a plan that stacks book after book onto the same sword drags a doubling penalty through every remaining step, while a plan that pairs books off first pays it once.

Combining books first is often worth an extra operation. Each enchantment costs its final level multiplied by a per-enchantment figure, and every operation also pays both sides' penalties. Folding two books into one before touching the gear therefore trades an extra step now for a cheaper one later — and whether that trade pays depends on the levels, the figures and the penalties involved, which is precisely why this page searches rather than following a rule.

The cost is the final level, not the increase. Putting a Sharpness V book onto a sword that already has Sharpness V still charges five levels times the multiplier, and changes nothing. That is the least obvious way to waste levels in an anvil, and the plan calls it out by name whenever a step does it.

Equal levels promote; unequal ones do not. Two Sharpness IV books make Sharpness V. A Sharpness III book on a Sharpness V sword makes nothing, and still costs.

Above 39 levels the anvil refuses. Outside Creative mode an operation worth more than 39 experience levels is not performed at all — the anvil shows "Too Expensive!" and there is nothing you can pay to get past it. That is why this page reports the worst single step separately from the total.

Levels, not experience points. Every figure here is a count of levels, which is what the anvil charges and what your screen shows. Minecraft levels are not evenly spaced in underlying experience, so this page does not convert between the two and never calls a level count "XP".

Most enchantment pages answer a different question from the one you have. They let you tick the enchantments you would like and total up a price for a set of perfect, freshly-written books nobody owns. You do not own that. You own three books from a librarian, one from a fishing trip, and a pickaxe you already put Efficiency on last week.

So this page starts from what is in your inventory. Every input has its own prior work state, because a book that has already been combined is not the same book. The item you are enchanting has one too, and it is usually the reason a plan that looks affordable is not.

The order is searched, not guessed. "Combine the cheap ones first" is a rule of thumb, and it is wrong often enough to matter: the penalty a pair leaves behind is paid again by every later step that touches it, so an operation that looks dear can be the cheapest thing you do. This page enumerates the ways the inputs can be paired and keeps the best, rather than following a rule.

And it will pay more to keep you under the limit. Given the choice between a cheaper plan with one refused operation and a dearer one you can actually carry out, it picks the one you can carry out, and tells you what the difference cost.

Java Edition only, and that is a finding rather than a gap. ToolGuild looked at whether the same rules hold in Bedrock Edition. Two differences are documented and could have been built: Bedrock charges only the increase in an enchantment's level rather than the final level, and does not add the one-level surcharge for a clashing enchantment. But nothing ToolGuild read establishes whether Bedrock refuses an operation above 39 levels — the source states the limit without naming an edition, and flags neighbouring text as unverified for Bedrock. "Too Expensive at step 4" is the most important sentence this page prints. Shipping it for an edition whose limit is not established would be worse than not shipping it, so the edition is not offered and this paragraph says why.

Nothing here rests on a Mojang statement, because ToolGuild did not find one. These are implementation-level mechanics, and the official material does not document them at this depth. The facts come from the Minecraft Wiki, the provenance line says "Community-derived" rather than borrowing an authority nothing supports, and the worked example below is one you can check in your own game in about two minutes.

You have a plain diamond sword, straight from the crafting table, and four fresh books: Sharpness V, Looting III, Unbreaking III and Mending.

The obvious thing is to put them on one at a time. Sharpness, then Looting, then Unbreaking, then Mending. That works, and it costs 29 levels, because by the fourth operation the sword has been through an anvil three times and is carrying a penalty of 7 before anything else is counted.

The plan this page returns costs 23, and the shape is the point:

1. Sword + Looting III book — 6 levels. Looting is one of the dearer enchantments to apply, so it goes on while the sword's penalty is still zero. 2. Unbreaking III book + Mending book — 2 levels. Two books, neither previously worked, so the whole cost is Mending at its book multiplier. 3. Sword + that combined book — 7 levels. One penalty on each side, then Unbreaking and Mending together. 4. Sword + Sharpness V book — 8 levels. Sharpness is the cheapest enchantment in the game to apply, so it is the one that pays the sword's accumulated penalty.

Six levels saved, and no single step above eight. Change one thing — say the sword has already been repaired twice — and the order changes with it, which is exactly why this is worth computing rather than memorising.

It covers Java Edition only. See the section above for why: the anvil's 39-level refusal is not established for Bedrock Edition in anything ToolGuild read, and that verdict is the most important thing this page prints.

It plans a target item and enchanted books. Combining two pieces of gear is a real anvil operation and Minecraft performs it happily — this page does not cost it, and the omission is deliberate. Doing it correctly needs the exact item on both sides, and "sword" here covers iron and diamond alike, which an anvil will not combine. It also needs durability, because an item-plus-item operation repairs the target too and the repair is part of what the anvil charges. Rather than guess at either, the page asks only for books.

It plans enchantments, not repairs. Repairing with raw material, and the unit costs that carries, are not modelled for the same reason.

It does not price renaming. Naming an item adds a level to an operation and does not itself accumulate a prior work penalty. Neither is in these figures.

It says nothing about how you get the levels. No experience rates, no farm throughput, no time estimates. It counts what the anvil charges.

Levels are not experience points, and this page does not convert between them. A 23-level plan has not cost "23 XP", and no figure here should be read as one.

The catalogue is a deliberate subset. A carved pumpkin, a head and a carrot on a stick are absent because nobody plans an enchantment order for one. That is a product decision, not a claim they cannot be enchanted.

Eight enchanted books is the limit. The search examines every way they can be paired, and that grows steeply. Eight covers every piece of gear ToolGuild could find a real use for; beyond it the page would be slow rather than wrong, so it declines instead.

It has never seen your inventory. Everything it knows is what you typed, including the prior work counts — which the game does not display, and which you may have to work out from what you remember doing.

The cost algorithm is community-derived, and the page says so. The Phase 3B.7 research gate did not find a first-party implementation-level specification of the anvil cost algorithm in the official surfaces checked, so the provenance line reads "Community-derived" and the verification level is 1. That is the honest label, not a disclaimer bolted on. It is also a narrower statement than it may look: Mojang’s own release notes do establish current catalogue facts, and one of them is cited above for the spear — but nothing first-party publishes how the anvil prices an operation.

In what order should I combine my enchanted books? That is what the planner above works out, from the books you actually hold. There is no single right order — it depends on which enchantments they are, what is already on the item, and how many times each thing has been through an anvil.

Why does my anvil say "Too Expensive!"? Because that operation would cost more than 39 experience levels, and outside Creative mode the anvil refuses those outright. It is almost always the prior work penalty rather than the enchantment: after five anvil visits an item carries 31 levels of penalty before anything else is counted.

Can I get past "Too Expensive"? Sometimes, by doing the same combinations in a different order — which is what the planner searches for. Sometimes not, and then the page says so rather than showing you a plan you cannot carry out.

What is the prior work penalty? A hidden count of how many times something has been through an anvil. The levels it adds double each time: 0, 1, 3, 7, 15, 31. Both items in an operation pay theirs, and the result inherits the higher count plus one.

Does renaming an item increase the penalty? Renaming adds a level to the operation but does not by itself accumulate a prior work penalty. This page does not price renames.

Why does it combine books together before touching my gear? Because every operation charges both sides' prior work penalties as well as the enchantments, so folding two books into one can be cheaper overall than applying them separately — and sometimes it is not. That is the trade the search is for.

Can I combine two swords instead of using books? In Minecraft, yes. On this page, no: costing it needs the exact item on both sides — an iron sword and a diamond sword will not combine — and the durability of the target, because that operation repairs it as well. Neither is something this page knows, so it asks for books and says so rather than guessing.

I have a Sharpness book and a Smite book. Which does it use? Neither, until you say. They cannot both end up on one sword, and choosing between them is a decision about the sword you want rather than about cost — so the page asks you to remove one instead of picking for you.

Two Sharpness IV books — do I get Sharpness V? Yes. Equal levels promote by one, up to the enchantment's maximum. Unequal ones do not: the higher simply stands.

My book has an enchantment the item already has. Is it free? No. The anvil charges the final level times the multiplier whether or not anything changed. The plan names those steps explicitly.

Is the answer the same in Bedrock Edition? ToolGuild does not know, and does not pretend to. Bedrock's cost formula is documented and different; its 39-level refusal is not established in anything ToolGuild read. Only Java Edition is offered here.

Are these figures experience points? No. They are experience levels, which is what the anvil charges. The two are not proportional and this page does not convert between them.

Are my saved plans sent anywhere? No. They are kept in this browser, they have no account behind them, and "Start over" clears them. Each one stores what you entered, and the plan is recomputed when you load it — so a correction to the rules can never leave you looking at a stale answer.