1RM calculator: Epley one rep max with a percentage chart

Enter what you actually lifted and get an estimated one rep max, plus the working percentages you would use to program around it. No need to test a true max.

Updated: 2026-08-31

Calculator

How to calculate 1RM: the Epley formula

To calculate your one rep max, take a recent set done with good form and multiply the weight by (1 + reps / 30). That is the Epley formula, the most widely used estimate:

estimated 1RM = weight x (1 + reps / 30)

So a set of 100 kg for 5 reps estimates a one rep max of about 116.7 kg. You can do it on paper in a few seconds, which is worth knowing when you are standing in a gym without your phone.

This is the same formula AIm uses internally, so an estimate on this page matches what the app shows for a logged set.

What the 1RM percentage chart is for

Most published programs prescribe load as a percentage of your one rep max: five sets of three at 85 percent, or three sets of ten at 70. The table above converts your estimate into those working weights so you can follow such a program without ever testing a max.

Round to whatever your gym actually has. If the table says 87.5 kg and the smallest plates give you 85 or 90, take 85 and aim to beat the reps.

One caution on the rep column in that table. It is a typical correspondence, not a rule: how many reps you personally get at 80 percent depends on the lift and on how you train. Squats usually allow more reps at a given percentage than deadlifts do, and someone who lives in the 5 rep range will get fewer reps at 70 percent than someone who trains at 12.

Why estimate instead of testing

Testing a true one rep max is useful occasionally and costly often. It is a maximal effort with a real injury risk, it needs a spotter on some lifts, and it leaves you too fatigued to train properly for a few days afterwards. For most people not competing in a strength sport, there is no reason to do it more than rarely.

An estimate from a normal working set costs nothing and can be recalculated after every session. That makes it a better progress signal than a max test, because you get one every week instead of once a training block.

How accurate is it?

Good in the middle, unreliable at the edges. Between roughly 3 and 8 reps the estimate is usually close. Above about 10 reps it drifts high, because at that point your set is limited by muscular endurance and technique breakdown rather than pure strength. A set of 20 will produce a number you almost certainly cannot lift.

It is also individual. Two people with the same 5 rep performance can have genuinely different true maxes, depending on fibre type and how practised they are at heavy singles. Someone who trains mostly in the 8 to 12 range will usually under-perform their estimate on the day.

The practical framing: treat it as a way to compare your own sets over time, not as a prediction of what you would lift today. A rising estimate means you are getting stronger, which is the question that actually matters.

The Epley formula, Brzycki and Lombardi

Epley is not the only formula. The two other common ones are Brzycki and Lombardi:

• Epley: weight x (1 + reps / 30)
• Brzycki: weight / (1.0278 - 0.0278 x reps)
• Lombardi: weight x reps to the power of 0.10

At 5 reps they agree closely. Take 100 kg for 5: Epley gives 116.7, Brzycki 112.5, Lombardi 117.5. By 10 reps they have spread out, and past 12 the disagreement between them is larger than the difference a month of training would make.

Which means the choice matters less than the consistency. Pick one, use it for every set, and compare the number against your own past numbers rather than against someone else's. Switching formulas halfway through a training block will show you a jump or a drop that never happened in the gym.

Bench press, squat and deadlift max

The formula is the same for every lift, but its accuracy is not.

Bench press max

Bench press estimates are the most reliable of the three, because the set usually ends when the chest and triceps fail rather than anything else. A set of 5 taken close to failure gives a number you can plan against.

Squat max

Squat estimates run slightly optimistic from high rep sets, because legs handle high reps better than the formula assumes: you get more reps at a given percentage, so the arithmetic reads that endurance as extra strength. Estimate from 3 to 6 reps and the drift mostly disappears.

Deadlift max

Deadlift estimates drift the most: grip and lower back fatigue end a long set early, so a 10 rep pull understates what you could lift once. This is the lift where staying in the low rep range matters.

The practical rule: estimate from sets of 3 to 6 reps where you can, especially on the deadlift, and compare estimates from the same rep range over time. An estimate from your 5 rep squat this month against your 5 rep squat last month is a clean comparison. Against your 12 rep squat it is not.

If you log RPE

RPE folds into the same formula through reps in reserve. RPE 8 means you had about 2 reps left in the tank, RPE 9 about 1. Add those reps to the reps you actually did and enter the total in the calculator above: 100 kg for 5 at RPE 8 counts as a 7 rep effort, which estimates about 123 kg.

This is an estimate stacked on an estimate, since RPE itself is a judgement call, so expect a wider error than from a set taken close to failure. The trade is worth it when you want a max number without grinding out the last reps.

Using it to drive progression

The reason an estimated max is worth tracking at all is that it makes different sets comparable. Consider two weeks:

• Week one: 100 kg x 5. Estimate about 116.7.
• Week two: 95 kg x 8. Estimate about 120.

The bar was lighter in week two and the performance was better. Judged by weight alone you would think you had gone backwards. This is the single most common reason people believe they have stalled when they have not.

The progressive overload guide covers what to change when the trend really has flattened.

Tracking it by hand across every exercise gets old fast. If you would rather not, describe the set to an AI assistant and let it keep the log:

With AIm connected, the assistant records the session and answers from your actual history: the estimate for that lift and its trend over the last weeks.