Mean calculator
Add up a list and divide by how many numbers it holds: 2, 4, 6, 8 and 10 sum to 30, so the mean is 6. Every value counts equally, and a stray label is flagged by its position instead of being quietly skipped. Free in your browser.
- Free
- Runs in your browser
- No sign-up
- Your data stays on your device
Run — free
Worked example
- Numbers
- 2, 4, 6, 8, 10
- Quantity
- 5
- Sum
- 30
- Mean
- 6
Students, tutors, and analysts reach for a mean calculator when they already have a finite sample and they want the sample mean — the value textbooks write as x-bar — without opening a spreadsheet. This calculator accepts a non-empty list of finite numbers and returns the count, the parsed values, the sum, and the arithmetic mean together with the formula used.
What the arithmetic mean is and when the sample mean is the right summary
The arithmetic mean of a list of numbers is the sample mean: add every observation and divide by the count n. If the list is two, four, six, eight and ten, the sum is thirty and the mean is six. That is the number people want when they search for a mean calculator: one representative value that treats each observation with equal weight, the quantity textbooks write as x-bar equals the sum of x-sub-i over n.
The formula does not apply weights, does not take a geometric product, does not invert reciprocals, and does not report the median or the mode. It is the right tool when every number already lives on the same scale and deserves the same influence: exam scores that are not yet percents, daily temperatures in one city, unit prices of comparable items, sensor readings from one instrument, or a short column of measurements copied from a notebook. It is the wrong tool when some observations should count more than others.
Combining a class of ten scores with a class of a thousand hides the large group. Negative numbers and zeros are accepted because they are still finite. Empty lists are rejected: the mean of an empty sample is undefined in this contract.
Any entry that is not a finite number is rejected by index rather than skipped, so a stray label, NaN, or infinity cannot silently shrink the sample and produce a fake mean. A single-element list is valid: the mean of one number is that number itself.
Where a mean calculator earns its keep, and what it is not
Use this mean calculator when you already have a list of numbers and you want their unweighted arithmetic mean, the statistic classrooms call the sample mean. Teachers compute the mean of raw quiz points that are not yet percents. Analysts compute the mean of daily temperatures, unit prices, or sensor readings from one instrument.
Spreadsheet users paste a column and want a second opinion that does not depend on a hidden filter or a silent skip of blank cells. Prefer a weighted mean when sample sizes or durations differ a lot. Prefer the median when a few outliers would drag the arithmetic mean away from the typical value.
Pin the rounded mean field in tests. Round-trip checks are easy: add the numbers by hand, divide by the count, and confirm the ratio matches the published mean within twelve decimal places.
What you can do with it
Lab measurements and sensor readings
Summarize a column of temperatures, prices, or instrument readings into one unweighted arithmetic mean for a daily report.
Notebook paste and independent check
Paste a short numeric column and confirm the mean independently of a hidden filter or a silent skip of blank cells.
FAQ
What happens if the list is empty or a value is not a finite number?
The mean of an empty list is undefined, and NaN, infinities, and non-numeric entries are not skipped. Failed validation is not billed.
Does this calculator weight some numbers more than others?
No. Every value has equal weight. If observations should count differently, use a weighted mean and supply the weights yourself.
Can the list include negatives or zero?
Yes. Negative numbers and zeros are accepted as long as they are finite. A single-element list is also valid: the mean of one number is that number.
Can I paste a column of numbers from a spreadsheet?
Yes. Figures that arrive as text are read as numbers, so a pasted column works. If one entry is a word or a stray label, it is flagged by its position instead of being dropped, so the count stays honest.
When is the arithmetic mean the wrong average?
When some values should count more than others. Averaging the mean score of a class of ten with that of a class of a thousand hides the larger group; a weighted mean is the honest summary there.
Related tools
For developers — API access
Everything on this page is available programmatically. This section is for teams who want to wire it into their own systems; everyone else can just use the tool above.
API endpoint
Prefer to automate it? One authenticated POST creates the task; the result comes back by webhook or a signed link. The same capability also runs here on the web, by email and from Telegram — and soon from our app too.
Call it from your stack
curl -X POST https://api.kit.forhosting.com/math/mean \
-H "Authorization: Bearer $KIT_KEY" \
-H "Content-Type: application/json" \
-d '{"numbers":[2,4,6,8,10]}'const res = await fetch("https://api.kit.forhosting.com/math/mean", {
method: "POST",
headers: {
"Authorization": `Bearer ${process.env.KIT_KEY}`,
"Content-Type": "application/json"
},
body: JSON.stringify({
"numbers": [
2,
4,
6,
8,
10
]
})
});
const { task_id } = await res.json();import os, requests
res = requests.post(
"https://api.kit.forhosting.com/math/mean",
headers={"Authorization": f"Bearer {os.environ['KIT_KEY']}"},
json={
"numbers": [
2,
4,
6,
8,
10
]
},
)
task_id = res.json()["task_id"]<?php
$res = file_get_contents("https://api.kit.forhosting.com/math/mean", false, stream_context_create([
"http" => [
"method" => "POST",
"header" => "Authorization: Bearer " . getenv("KIT_KEY") . "\r\nContent-Type: application/json",
"content" => '{"numbers":[2,4,6,8,10]}',
],
]));
$task = json_decode($res, true);body := bytes.NewBufferString(`{"numbers":[2,4,6,8,10]}`)
req, _ := http.NewRequest("POST", "https://api.kit.forhosting.com/math/mean", body)
req.Header.Set("Authorization", "Bearer "+os.Getenv("KIT_KEY"))
req.Header.Set("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)Example request
{
"numbers": [
2,
4,
6,
8,
10
]
}Example response
{
"task_id": "tsk_a1b2c3d4e5f6a1b2c3d4e5f6",
"type": "math.mean",
"status": "queued",
"_links": {
"result": "/tasks/tsk_…/result"
}
}The API is asynchronous: the call returns a task_id immediately and the result arrives by webhook. Polling is capped at 1 req/s per task.
Pricing
Published price — no tokens, no invented credits. A failed task is never charged.
Limits
max_items | 100000 |
Errors
| HTTP | Code | Meaning |
|---|---|---|
401 | unauthorized | Missing or invalid API key. |
402 | insufficient_balance | Your balance doesn't cover the task price. |
404 | unknown_type | That task type doesn't exist. |
429 | rate_limited | Too many requests. Use the webhook instead of polling. |
Technical notes
The arithmetic mean is the statistical centre of a list of numbers: add every observation and divide by how many observations there are. An empty list or any value that is not a finite number, including NaN and infinities, fails with invalid_input and is never billed. A successful API call costs $0.002.
What the arithmetic mean is and when the sample mean is the right summary
In that case a weighted mean is the honest summary, and this endpoint will not invent weights you did not supply. Magnitudes above one quadrillion are also rejected so the Worker and the browser stay inside a safe numeric range.
Input contract, response shape, and error behaviour
Numeric strings inside the array are coerced, so a pasted column of figures still works. On success the body reports count as n, numbers as the parsed list in input order, sum as the total of those values, mean as sum divided by n, and a formula string restating (x1 + x2 + … + xn) / n. A single-element list is valid and returns that number as the mean. Send a JSON body with one required field named numbers: a non-empty array of finite numbers. Aliases values, list and data are accepted and behave the same. Failed validation — missing numbers, an empty list, a non-finite entry such as NaN or Infinity, a list longer than one hundred thousand values, or a magnitude above one quadrillion — returns invalid_input naming the problem and is never charged. A successful API request is charged the published $0.002 base price with unit item. Pair classroom examples deliberately: two four six eight ten produce six; one point five, two point five and negative one produce one; an empty list demonstrates the invalid_input path that refuses to invent a mean.
Where a mean calculator earns its keep, and what it is not
Pair classroom examples deliberately so learners can check by hand: two, four, six, eight and ten produce six; ten, twenty and thirty produce twenty; an empty list demonstrates the invalid_input path. This page is not a mean-median-mode summary, which reports three descriptive statistics at once and ranks for a different search. It is not an average-percentage calculator, which treats the figures as percents already expressed out of one hundred. It is not a weighted mean, a geometric mean, or a harmonic mean. Keep magnitudes inside the published absolute limit and keep the list within one hundred thousand values so interactive Worker and browser paths stay fast. Failed tasks that return invalid_input do not consume prepaid balance, which keeps exploratory scripts cheap when learners send empty arrays or stray text. API pricing remains the flat $0.002 per successful request.
Integration use cases
Exam scores and homework checks
Compute the sample mean of a list of raw scores (2, 4, 6, 8, 10) and check x-bar against a deterministic browser or API result without converting to percents first.
Technical questions
How is the mean calculated?
Send a non-empty list of finite numbers. The API returns (x1 + x2 + … + xn) / n as mean, plus count and the sum.
What does the mean calculator cost?
The browser tool is free. API requests use prepaid KIT balance at $0.002 per successful call; invalid input is not billed.