These are ranked strictly by stated tolerance, because that is what the question asks. But the tolerance on the box is the smallest error in your measurement, and it is worth knowing what the bigger ones are before you spend money on a tighter number.
We earn a commission if you buy through the links below, at no extra cost to you. It never changes what gets recommended or how these are ranked.
| # | Best for | Model | Stated tolerance | Range | Rating | |
|---|---|---|---|---|---|---|
| 1 | Tightest spec | Leica Disto D5 | ±1/25 in | 650 ft | 4.3 · 75 | Check price |
| 2 | Jobsite | Leica Disto X3 | ±1/25 in | 500 ft | 4.1 · 76 | Check price |
| 3 | Value | Bosch GLM165-40 | ±1/16 in | 165 ft | 4.7 · 5,846 | Check price |
| 4 | At range | Bosch GLM400CL | ±1/16 in | 400 ft | 4.6 · 933 | Check price |
| 5 | Layout | Stabila LD250BT | ±1/16 in | Not listed | 4.5 · 108 | Check price |
| 6 | MEP work | Fluke 417D | ±2 mm | 131 ft | 4.5 · 157 | Check price |
Tolerances are the manufacturers' stated figures for a single measurement under favourable conditions. Roughly, 1/25 in is one millimetre and 1/16 in is 1.6 millimetres.
Every unit in this guide is honest about its specification, and the gap between the best and the worst of them is about one millimetre. Here is what that one millimetre is competing against.
A laser measure can reference from its front edge, its back edge, or from a folded out end piece. Owners of the Bosch GLM400CL note that it defaults to measuring from the back of the device. That is a sensible default when you are pushing the tool against a wall, and completely wrong when you are holding it out over an edge.
These units are around five inches long. Get the reference end wrong and you are out by five inches. Against a tolerance of one sixteenth of an inch, that is an error eighty times larger than the spec you paid for. Nothing else on this page comes close to costing you that much.
One Huepar owner put it precisely. Accuracy was good, roughly a sixteenth over sixteen feet, but it is much harder to position than a tape measure, there is no clip to hook over an edge, and it is hard to press the measure button without causing the device to shift.
That is the real world error on every handheld measurement anyone takes. A Leica D1 owner reached the same conclusion from the other direction, finding it hard to hold and get a good result and wondering whether they needed a pole. They probably did. A tripod thread or a stable edge to brace against will do more for your numbers than moving from one sixteenth to one twenty fifth.
This is the one that surprises people. If you are calculating an indirect height, the tilt sensor's tolerance dominates the result, not the distance sensor's.
Bosch quotes levelling accuracy of plus or minus 0.2 degrees on Blaze units. Take a 50 ft shot at 40 degrees to work out a height. Two tenths of a degree of angle error moves the calculated height by about 1.6 inches. The distance tolerance over that same shot contributes a sixteenth of an inch. The angle error is roughly twenty five times larger.
So if your work involves heights you cannot reach, the number to care about is the levelling accuracy, and the thing to do is check the tilt sensor against a spirit level rather than shopping for a tighter distance spec.
A residential appraiser using the GLM400CL reported that it will occasionally present the previous reading as though it were a new one, rather than showing a failed read. An obviously wrong number gets caught. A plausible number that belongs to the last wall you measured does not. If a reading appears instantly and matches something you took a moment ago, take it again.
Manufacturers quote tolerance for a single distance measurement to a favourable target in favourable light. Three things fall outside it.
Derived values compound. Area and volume multiply two or three measurements together, so they carry more error than any one of them. A room measured to a sixteenth in both directions is not a floor area accurate to a sixteenth.
Poor targets widen it. Dark, glossy, wet or angled surfaces all reduce the strength of the returning signal, and the quoted figure assumes none of them. This is separate from range, which is the more familiar version of the same problem.
Units matter as much as tolerance. This is why the Leica model number warning below is on every page we publish about that tool. A unit that will only show decimal feet is useless for framing regardless of how tight its specification is, because you will be doing the conversion in your head and rounding as you go.
The US Disto X3 is model 850834. Buyers repeatedly receive 850833, the international version, which cannot display sixteenths of an inch. On a page about accuracy this matters more than anywhere else. The tightest tolerance in this guide is worth nothing if the display rounds to decimals. The box tells you which one you have.
The tightest stated tolerance here, and the only unit that also solves the aiming problem that turns a good spec into a bad measurement at distance.
Right for: surveyors and estimators where a millimetre genuinely matters and the shots are long.
Read our full review of the Leica Disto D5
Check price on AmazonThe same tolerance as the D5 in a body that will survive being dropped, which is a form of accuracy nobody puts on a spec sheet. A tool that has been knocked about holds its calibration better when it was built to be knocked about.
Right for: precision work that happens on a live site rather than in a finished building.
Read our full review of the Leica Disto X3
Check price on AmazonFor most trades this is where the accuracy conversation should end. A sixteenth of an inch is the unit the work is specified in, and almost six thousand ratings say it holds it.
Right for: anyone whose drawings are dimensioned in sixteenths, which is most people reading this.
Read our full review of the Bosch GLM165-40 Blaze Pro
Check price on AmazonHolds a sixteenth out to 400 ft, which no other unit at this price claims. More importantly, the camera means the reading you take is of the thing you meant to measure.
Right for: long interior shots and outdoor estimating where aiming, not the sensor, is the error source.
Read our full review of the Bosch GLM400CL Blaze Outdoor
Check price on AmazonA sixteenth of an inch from a company whose entire reputation rests on things being straight. The hardware is not in question here. The software and the documentation are.
Right for: setting out work where dimensions get transferred rather than read once.
Read our full review of the Stabila LD250BT
Check price on AmazonLast on tolerance and it is the only honest place to put it. Fluke publishes plus or minus 2 mm, looser than everything above. For running conduit that has never once mattered, and the tool is well liked for exactly that reason.
Right for: electricians and mechanical trades who want one button and a Fluke warranty.
Read our full review of the Fluke 417D
Check price on AmazonFor most people, no. The step from one sixteenth to one twenty fifth of an inch costs several times the price of the tool and is smaller than the error you introduce by holding the unit at arm's length. If you are framing, tiling, hanging cabinets or estimating materials, the GLM165-40 is the accurate laser measure you are looking for and the two Leicas above it are solving a different problem.
The millimetre is worth paying for in three situations. When the measurement is going into a document somebody else relies on, such as an appraisal or a survey. When you are calculating rather than reading, because derived values compound their inputs. And when the tool has to keep its calibration through years of abuse, which is the X3 argument rather than a specification argument.
A Makita LD050P was removed after an owner reported never being able to get an accurate measure out of it, which on a page ranking accuracy is disqualifying regardless of the brand on the case.
A budget unit advertising a 393 ft green beam was removed after owners measured errors of three quarters to a full inch across a 60 inch span. That is not a tolerance problem, it is a different category of tool pretending to be this one.
Several sub fifty dollar units were excluded for the same reason. They are covered honestly in the budget guide, where the gap between claimed and delivered performance is the whole subject.
Every unit here ships within its stated tolerance and some do not stay there. Check yours against a known length on the day it arrives, and check the tilt sensor against a spirit level if you intend to use indirect height at all. The procedure takes five minutes and it is the only way to know which of these numbers your particular unit is actually holding.
You will find our overall picks on the front page, with the reasoning behind each one.