Every technique you need to solve a Light Up grid by logic alone — no guessing.
Akari tips are really about one thing: learning where to look. A Light Up puzzle has exactly one solution, and there is always a next move that follows purely from the rules — you just have to notice it. The strategies below are ordered by when you'll actually need them, from the number-cell reads that crack a board open in the first ten seconds to the elimination chains that finish the hardest 14×14 Weekend Super grids.
Start with whichever one matches the number staring at you. Together they handle the overwhelming majority of positions you'll meet:
A number 0 cell tells you that none of its four adjacent cells (up, down, left, right) can hold a bulb. It is the cheapest information on the board — the moment you see one, mark all four neighbours with X and move on. Every X you place shrinks the candidate space for the numbers around it.
Zeroes also matter later in the solve: any bulb placement you are considering must not touch a 0's neighbours. When a region seems to have two possible solutions, check whether one of them lights a cell adjacent to a 0 — if it does, that solution is dead and the other one is forced.
A number 1 cell means exactly one of its four adjacent cells holds a bulb — no more, no less. If three of its neighbours are already blocked (by walls, X marks, or the grid edge), the remaining neighbour must be a bulb. That forced placement is free progress.
The harder read is line-of-sight: once you know one specific neighbour is the bulb, the other three are X'd out — and if placing that bulb would illuminate a cell that already holds a bulb, your read is wrong somewhere. Trace back before you place anything else.
A number 2 cell needs exactly two of its four neighbours lit. Corners and edges are where 2s become powerful: a 2 in a corner has only two available neighbours, so both must be bulbs — a forced placement, no deduction needed. A 2 on an edge with one neighbour blocked has exactly three candidates and usually narrows fast.
In the middle of the board a 2 carries less information on its own, but it compounds: every neighbouring cell you X out turns it one step closer to forced. Track 2s alongside their neighbours rather than in isolation.
A number 3 cell needs three of its four neighbours lit — only one is blocked. Flip the question around: instead of asking "which three get bulbs?", ask "which one can't?". If any neighbour is proven impossible (it would see another bulb, or it's next to a 0), the other three are all forced placements.
Threes are the second-richest read after 4s. On GridPaw boards a 3 squeezed between walls usually cracks open a whole corner of the grid, because its three forced bulbs illuminate two full lines each.
A number 4 cell means all four adjacent cells must contain bulbs — there is nothing to deduce. Place all four the moment you spot one, and check the consequences: each of those bulbs illuminates its own row and column, so before making any other move, see which nearby numbers just became forced or blocked.
The only thing a 4 can't tell you is where its bulbs' light must stop — that's what the walls decide. After placing a 4's bulbs, re-scan the numbers touching those four lit lines; a 1 next to a newly lit neighbour often becomes a forced X chain.
The single biggest beginner mistake in Akari is placing bulbs reactively — spotting an obvious one and clicking, then hunting for the next. Instead, do a full sweep of the grid first and read every number. A number 0 cell instantly tells you four cells are dead. A number 4 cell hands you four bulbs. Those two reads alone often account for a third of a puzzle's bulbs, and they cost you nothing but a single pass with your eyes.
This is the same discipline a solver automates. When you ask an Akari solver for a hint, what it returns first is almost always one of these forced placements — because forced placements are provable, and provable moves don't need search.
Every bulb illuminates its own cell and every cell along its row and column in both directions, stopping at the first black wall. Two consequences matter constantly:
Once line of sight is automatic, the elimination technique becomes natural: instead of asking "where can a bulb go?", ask "what breaks if I put one here?"
New solvers tend to scan for empty cells and wonder what belongs there. Strong solvers scan the number cells and the walls. Numbers and walls are the only things on the board that carry information — empty cells don't. This habit is the difference between a two-minute solve and a stalled one, and it's the same reason advanced Akari boards feel like deduction rather than search.
A 7×7 easy grid is usually 1–3 minutes. A 10×10 daily puzzle takes most solvers 5–10 minutes. The 14×14 Weekend Super Challenge can run 20 minutes or more if you're being careful.
No. Every GridPaw puzzle is verified to have a unique solution reachable by pure deduction. Guessing means you missed a constraint read — usually a number cell you skimmed or a wall shadow you didn't account for.
Read all the 0s and 4s before you touch anything else. Those two numbers give you free, provable placements, and they're the fastest way to turn an intimidating grid into a mostly-solved one.
Start with Easy Akari Puzzles to build the number reads, then move to the Daily Challenge where the grid is bigger and the elimination chains get longer.
If you'd rather see the rules from the top before drilling technique, the How to Play guide covers the basics, and the solving walkthrough works through a full grid step by step.
