How seed finding works (and how to use SeedForge)

Every generated world starts from a seed — a 64-bit number that feeds the world generator. The same seed always produces the same terrain, the same biomes and the same structure layout. That determinism is what makes seed finding possible: instead of exploring worlds one by one, a seed finder runs the generator math for millions of seeds and keeps the ones that match your wishlist.

Why this can run in your browser

SeedForge compiles the world-generation library to WebAssembly and runs it in parallel across your CPU cores. Structure positions are decided by a cheap random-number check per region, so most seeds can be rejected in nanoseconds — your machine can test hundreds of thousands of seeds per second for structure searches, entirely offline. Nothing is uploaded; the search runs on your device.

Building your first search

1. Open the seed finder and pick your game version — world generation changes between versions, so a seed found for 1.21 will look different in 1.16.
2. Add a condition. Start simple: Village within 500 blocks of spawn.
3. Press Start. Results stream in live; click any row to see that seed on the map.

Chaining conditions

The real power is chaining: each condition can search relative to where a previous condition was found. For example:

• Village within 500 of spawn, and
• Ruined Portal within 256 of that village, and
• Ocean Monument within 300 of the portal.

Chained searches multiply in rarity — the difficulty estimate under the condition list warns you before you start a search that could take hours.

Reading the difficulty estimate

The estimate is a rough statistical model: it multiplies the probability of each condition to guess how many seeds must be checked per match. "1 in 300" finds dozens of matches per second; "1 in 5,000,000" may take many minutes. Radius is the biggest lever — doubling a radius roughly quadruples the odds.

Ready? Open the seed finder →