Minecraft Speedrun Seeds: 11 Distances in 33 Pages, and Only One Page With a Budget

How we counted, before you read any number on this page

Every figure below comes from scanning this site's own pages, not from the game. The rule comes first so you can check us.

What counts as a limit-word statement. A figure written in digits, followed by blocks, with a limit word immediately in front of it: within, over, under, less than, more than, at least, no more than, beyond, further than. Within 500 blocks counts. Over 3,000 blocks counts. That rule returns 39 statements across 15 pages, and those 39 hold seven distinct numbers: 10, 200, 256, 500, 600, 1000 and 3000.

What is a distance but not a limit-word statement. About 27 blocks away, eleven blocks down, 300 blocks west, about 388 blocks of Nether travel. Every one of those is a real distance and every one is in the table below, but none has a limit word in front of it, so none is counted as a statement above. They are four more numbers, and they are the four that make the table eleven rows instead of seven. We list them rather than drop them, because two of the four are the most useful distances on the site.

What does not count. Coordinates on their own, seed codes, version numbers, Y levels, and anything in a navigation bar or a footer.

What a record is. Forty-nine of them: twenty rows in the coordinate table and twenty-nine archive cards from 2011 to 2019. The scan covers the pages, which is where the distance writing lives; the records are what the numbers are about.

The thing we are not doing. We are not estimating how long anything takes in someone else's run. The only clock number on this site is the one attached to 2016000, it was logged by the tester who made it, and we are not generalising from it.

The eleven distances: seven behind a limit word, four that describe one seed

Thirty-nine statements, fifteen pages, seven numbers behind a limit word. The table lists all eleven distances we publish, because the four without a limit word are not less useful — a village 27 blocks out and a stronghold 11 blocks down are the reason the only timed run we hold finished in 27 minutes. Read down the column and the shape of the site becomes obvious: most of these numbers describe one seed once and are never used again.

Four of the eleven are budgets you can apply to a seed you have not loaded yet: 200, 256, 600 and 1000. Three more sit behind a limit word without being budgets — 10 and 500 are measurements, and 3,000 is a rejection verdict. One is a conversion. The rest describe a single record once.

The only page with a real budget

One page on this site asks four distance questions at the same time and requires all four to pass. It is the server-grade standard, and the numbers are: a witch hut within 256 blocks, an ocean monument within 1000 blocks, a pillager outpost within 600 blocks, and a second warm ocean or a second monument in another direction.

The reason all four are required matters before you borrow the numbers, because it explains why this is not a speedrun budget. A multiplayer server is four players spending the same first night, and any single missing structure forces a longer walk for whoever lands in that quadrant. The budget exists to make the worst case survivable for the group, not to make the best case fast for one person.

That difference is the whole gap between what this site has and what a speedrun seed needs. A server budget asks is anything too far. A speedrun budget asks is everything close enough, in order, on a clock. Ours answers the first question and has never been pointed at the second.

388 blocks: the one number that thinks in travel time

A quad ocean monument sits at -3100, -1100. On the surface that is over three thousand blocks from spawn, which on its own is a disqualifying distance by any standard on this site. It survived review for one reason: the tester noted that Nether travel compresses the trip to about 388 blocks.

This is the single most speedrun-shaped sentence in our records, because it is the only place where we stopped measuring distance in blocks and started measuring it in travel time. It is also the only place where one trip carries two honest numbers — three thousand and three hundred and eighty-eight — and both are true.

The catch is the one a speedrunner will spot immediately: the compression only exists once the highway does. Four hundred blocks of Nether travel is a short hop after the portal and the tunnel are built, and a building project before that. The number is real and the discount is conditional.

The number that got a seed thrown out

Seed -9147459640944998301 on Java 1.18+ holds a mushroom island, a quad ocean monument, a witch hut and a pillager outpost. By the contents test it is a headline seed. The record says it was rejected anyway, and it says why in the tester's own words: slightly over one thousand blocks.

This is the only time on this site that a distance number changed the fate of a record rather than decorating it. It is the clearest evidence we have that distance is not a detail: the same features at a different spacing would have made the list, and the spacing is the only thing that was measured.

Compare it with 4188583766222650532, which carries nine coordinates, two Nether fortresses, a bastion and a mushroom island and stayed in. Same kind of haul, different spread, opposite outcome. The difference is written down, and it is the difference between a seed you can route and a seed you have to commute to.

The closest thing to a speedrun seed we have

Which Clock Are You Racing?

A distance is not a time. Before any number below is worth carrying into a run, you have to know which category you are running, because the category decides what that number costs you.

Three ways to start a world

The same walk to a temple is three different facts depending on how the world was created:

This is why -9147459640944998301 reads two ways. On this site it was rejected for slightly over one thousand blocks. In a random-seed run that sentence is a reset. In a set-seed run it is a route note: a boat, a stack of wood, and a decision about whether a quad monument is worth the crossing. Rejected: over distance Quad monument Readable two ways

The version bucket comes before the seed

Timed runs are split into four version buckets. Each one has a version people actually play, and none of them is the newest version on your launcher:

So a seed list without a version is not a run list. Our own table is mixed: the record with the timed run is Java 1.20–1.21, the record with nine coordinates is marked 1.18+ with no upper bound, and the most detailed Bedrock card is recorded as not working on Java at all. Copy the code without the version and you are comparing two different worlds.

Same code, different route

Three strategies in the guide are tagged by edition, and the tags are the part island lists never print:

• Village strategy — Bedrock only. It works because strongholds generate under villages in Bedrock: throw an eye, walk that way to a village, stand in the middle of it and dig down.
• Monument strategy — Java only.
• Insomniac strategy — Java only.

That is the exact shape of our 2016000 record on Java: village about 27 blocks from spawn, stronghold eleven blocks under the village well, portal lit at 27 minutes. It is a good island seed, and it is also, structurally, a Bedrock route that happens to exist on Java. The same number on Bedrock 1.20.5 drops you in open ocean at 0, 62, 0 with no trees, and the nearest land is a mushroom island 300 blocks west. Two records, one code, two games.

What the clock measures

The guide mentions in-game time once, as a progress check on one route: your IGT, or in-game time, should be around six to eight minutes when you reach a known point. It never states when the clock starts or stops, because those rules belong to the category, not to the world. A page like this one therefore cannot tell you a run is fast. It can only tell you what the ground will charge you before the clock matters: 27 blocks to a village, eleven blocks down to a stronghold, 525 blocks to a jungle temple, 388 blocks of Nether travel once a highway exists. Judge the seed on those, then judge the run on the rules.

Five checks before you call any seed a speedrun seed

1. Ask for the number, not the word. Of our forty-nine records, the ones that say nearby almost never say how near. If a list will not give you a figure with a unit, you cannot route it.
2. Check whether a budget is written down anywhere. Ours is written on exactly one page, the server-grade standard, and it was built for four players. A page with no stated budget has no standard it can fail.
3. Find out which distance decided the seed. On -9147459640944998301 it was slightly over one thousand blocks. That one number overwrote a mushroom island and a quad monument. Ask what the equivalent number is on the seed you are looking at.
4. Separate surface distance from travel distance. Three thousand blocks and 388 blocks are the same trip on our server-grade page. The second number is only available after the highway exists.
5. Check the edition before you copy the code. 2016000 is a twenty-seven-block walk to a village on Java and open ocean with no trees on Bedrock. One number, two generators.

Running these with other people

The four-number budget on the server-grade page is the right shape for a group and the wrong shape for a clock. Seed 4188583766222650532 is the one that works for both: nine coordinates, two fortresses, a bastion and a mushroom island close enough that one camp can service several of them. Seed -9147459640944998301 is the one that does not: the same haul, spread slightly over a thousand blocks, rejected by the person who measured it.

If you want to run one of these with other people on Java or Bedrock, Shockbyte is a low-cost host that sets up in minutes — use code LAUNCH for 25% off your first month. If you sign up through that link, I may receive a commission at no extra cost to you. It does not change what you pay.