# Root Mean Squared Error (RMSE)

> An error score for number guesses that punishes big misses more. Lower is better.

**Field:** Artificial Intelligence  
**Category:** Evaluation & Benchmarks  
**Also known as:** RMSE, Mean Squared Error (MSE), Root Mean Square Error

## Think of it like

It is like a dart game where far misses cost extra. One wild throw hurts a lot.

## Definition

Root mean squared error, or RMSE, scores a model's guesses on numbers. You take each gap between a guess and the true value and square it, so big gaps grow much bigger. The average of those squares is the mean squared error, or MSE. Then you take the square root to get back to real units. Because of the squaring, a few big misses push RMSE up a lot. It is always at least as big as MAE on the same guesses.

## Formula

```
RMSE = square root of the average of (true value - guess)²
```

## Example

Two weather models both have an MAE of 2 degrees. One misses by 10 degrees on a few days, so its RMSE is much higher. Emma picks the other one.

## Why it matters

When one huge miss is worse than many small ones, RMSE shows it. Many models are trained to make the squared error small.

## Common mistake

Reading RMSE as the usual miss. A few outliers can make it much bigger than most of the errors.

## Three principles

- **Square the gap.** Big misses grow fast.
- **Root brings units back.** It reads in real units again.
- **MSE inside.** RMSE is the square root of MSE.

## Related terms

- [Mean Absolute Error (MAE)](https://learn.benibix.com/term/ai/mean-absolute-error)
- [R-Squared](https://learn.benibix.com/term/ai/r-squared)
- [Loss Function](https://learn.benibix.com/term/ai/loss-function)
- [Outlier](https://learn.benibix.com/term/general/outlier)
- [Variance](https://learn.benibix.com/term/general/variance)
- [Regression (Machine Learning)](https://learn.benibix.com/term/ai/regression)

## Told apart from

- [MAE vs RMSE](https://learn.benibix.com/compare/mae-vs-rmse): People think RMSE is just another way to write the average miss, like MAE. It is not. MAE counts each miss by its size. RMSE squares each miss first, so one big miss pulls it up far more. When the two are far apart, a few wild misses are hiding in the data.

![Five dots sit near a straight line, each joined to it by a short gap. One dot sits far away. Three boxes show the scores: MAE, RMSE and R-squared.](https://learn.benibix.com/diagrams/ai5-regression-errors.svg)

---

Source: https://learn.benibix.com/term/ai/root-mean-squared-error
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-19
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
