Squaredle Solver: Find All Words in the Letter Grid

Squaredle Solver

Type the letters of your puzzle into this Squaredle solver and it finds every word you can trace by moving between touching squares: sideways, up and down, or diagonally. Results are grouped by length, and selecting a word lights up its path on the grid.

Type the letters, then press Find words. Words need at least 4 letters.

How do you solve a Squaredle puzzle?

You solve Squaredle by tracing words of at least four letters through squares that touch, diagonals included, without using any square twice in the same word. The solver does this for every possible path and keeps the ones that spell a word in the ENABLE word list. Use its list, longest words first, to find the words you have not spotted yet.

How to use the Squaredle helper

Pick the grid size: 3x3, 4x4 or 5x5.
Type the letters row by row, starting at the top left. The cursor jumps to the next square after each letter, and the arrow keys move it anywhere.
Leave a square empty if your puzzle has a hole there. You can also paste the whole grid at once, with a dot for each blank square.
Press Find words. The first search downloads the word list; after that, every grid is solved in a fraction of a second, even without a connection.
Select any word in the results to see its path: the squares are numbered in the order you would swipe them.

Squaredle rules in short

A Squaredle puzzle is a grid of letters. You form a word by moving from one square to a neighbouring square, and each square has up to eight neighbours because diagonal moves count. A word must be at least 4 letters long and may not visit the same square twice, although the same letter can appear in more than one square. Puzzles are not always the same size and may have blank squares, so this solver lets you choose the size and leave squares empty.

How the solver searches the grid

First, the solver drops every dictionary word that needs letters the grid does not have, or more copies of a letter than the grid holds. The remaining words go into a prefix tree, a structure that can tell instantly whether a sequence of letters is the start of any word. The solver then walks from each square to its neighbours, one letter at a time, and abandons a path as soon as it no longer matches the start of a word. Each finished word is recorded with the first path that spells it.

Because it checks every path, the list is complete for the word list it uses. That word list is ENABLE, not the puzzle's own list, so some results may not be accepted in the game and a few accepted words may be missing.

Example: a 3x3 grid

Enter the rows CAT / RSE / NOI. The solver finds 64 words of four letters or more. The longest are CRATES and ORATES, both six letters; CRATES runs C, R, A, T, E, S by stepping down, diagonally up, right, down and left. Shorter finds include AROSE, NOISE, ONSET. Even a small grid hides more words than most people find on their own, which makes the solver a handy Squaredle helper for checking what you missed.

Similar puzzles

The same search powers the Word Hunt solver, which is set up for a full 4x4 board and three-letter words. If your puzzle gives you a set of letters rather than a grid, the Spelling Bee solver is the better fit.

Frequently asked questions

How does a Squaredle solver work?

It searches the grid the way you would, but exhaustively: starting from every square, it follows touching squares in all eight directions, never reusing a square, and keeps every path that spells a word of four or more letters in its dictionary. It stops early on any path that cannot become a word, which keeps the search fast.

Can a word reuse the same letter?

A word can contain the same letter twice only if that letter sits in two different squares, because each square can be used once per word. The solver follows the same rule, so no path it shows visits a square twice.

What grid sizes are supported?

Square grids of 3x3, 4x4 and 5x5 are supported, and any square can be left blank. Pick the size from the menu and leave empty squares empty: the solver treats them as gaps that a word cannot pass through.