7 Free Reaction Time Tests, Compared

Quick answer: every reaction time test site uses the same core mechanic — tap when it turns green — but they differ in variant formats (like F1 starts), language support, and whether results feed into a ranking. Below are 7 notable ones, ranked by format.

Free reaction time tests exist all over the web, and the core mechanic is universal: the screen switches from red (wait) to green (go), and your tap-to-green delay is measured in milliseconds. What differs is variant formats, language support, and how your record is tracked. Here are 7 notable sites, described honestly — strengths and limits included.

Quick comparison

1. MIKIRI — Reaction Time Test & F1 Reaction Test

Alongside the standard reaction time test, MIKIRI offers an F1 Reaction Test that recreates Formula 1's start-lights procedure. Both support 17 languages and rate you S–D on a 5-try average. MIKIRI also has fifteen free games that isolate different speed-related skills (read-ahead reaction, rapid recognition, visual search, trajectory prediction), each with its own global top-100 leaderboard. No signup.

2. Human Benchmark

A long-running, well-known reaction time site. Alongside the standard test, it offers several other cognitive tests (memory, typing, etc.) and publishes a statistics page showing the global distribution of results.

3. Arealme

A large multi-quiz site offering a reaction time test among many others, localized into 34 languages — its standout feature is breadth of language coverage.

4. CPSTest.org

Offers the standard format plus variants including an "F1 Reaction Test" that mimics Formula 1 start lights, alongside click-speed (CPS) testing on the same site.

5. CPS-Check.com

Primarily a click-speed testing site that also offers a reaction time test — a simple, single-purpose toolset.

6. cpstest.us

A straightforward standard-format reaction time test tool.

7. cpstest.net

Offers both click-speed and reaction time testing tools.

Which one should you use?

For a one-off check of your reaction time, any of these will do — the core mechanic is the same everywhere. If you want multiple languages, an F1-style variant, and a broader set of games to compete on, MIKIRI is a practical pick. If you care more about large published statistics, Human Benchmark or Arealme are worth a look.

MIKIRI play data

This part is not general knowledge. These are values actually recorded on MIKIRI's server (as of 2026-10-04; each completed run counts once, so repeat plays are included; flagged runs excluded). The samples are small — tens to a few hundred runs per mode — so read them as where real runs tend to land, not as settled statistics.

Full distributions for every mode are on the statistics page. These are in-game results, not medical or academic measurements.

Why two tests give you two different numbers

Take the same test on two sites and you will almost always get two results. Neither is wrong; they are measuring slightly different things in slightly different ways. Before comparing, check these:

• Simple or choice reaction. "Press when it turns green" and "press only for green" are different tasks, and the second is slower.
• Number of trials and outlier handling. An average of five, a median, or a best-of — and whether false starts are discarded — all change the headline number.
• How the timer is taken. In a browser, the moment the stimulus is drawn and the moment the input event arrives both carry error. Assume every site has a systematic offset of a few to a few tens of milliseconds.
• Your device. The same site on a 60Hz laptop with a Bluetooth mouse and on a 120Hz monitor with a wired mouse will not agree.

So "230ms on one site and 260ms on another, so the first is more accurate" does not hold up. Pick the test whose method is stated clearly — trial count, how outliers are treated — and keep using that one on the same device.

What to look for in a test before trusting it

Instead of asking which site is "accurate", ask which one lets you measure the same way every time:

1. Does it randomise the delay before the signal? A fixed delay can be anticipated, and you end up measuring your timing, not your reaction.
2. Does it penalise or discard false starts? Without that, guessing early improves the score.
3. Does it show individual trials? A single averaged number hides one slow attempt that dragged it up.
4. Does it say what it measures? Simple reaction, choice reaction and movement-plus-aim are different skills and deserve different expectations.

MIKIRI's own reaction time test randomises the wait, rejects early presses and shows each trial, so that you can see the spread rather than just one number.

FAQ

Are all reaction time test sites equally accurate?

No. Browser implementation, display refresh rate, and touch latency can shift results by tens of milliseconds. Rather than comparing across sites, compare your own results on the same site and device over time.

Which ones are free and show rankings?

MIKIRI, Human Benchmark, and Arealme all publish free statistics or ranking-style information.

Where can I try an F1-style reaction test?

MIKIRI's F1 Reaction Test and CPSTest.org's F1-format test both offer this.

This article describes general features of third-party services at the time of writing; it is not a medical or academic endorsement of any specific service.

Related guides

• Reaction Time: What's Average, How Age Affects It, and How to Improve
• Dynamic Visual Acuity vs Vision Training: What's the Difference?
• How to Improve Your Reflexes (Honestly)
• Mobile vision training you can do on your phone
• Reflex games for kids and students
• Best dynamic vision games
• Sports vision training
• How to train dynamic vision
• Does brain training work? What the research does and doesn't show
• What science says about dementia prevention — 14 risk factors and the latest trials
• Forgetfulness vs dementia — the difference, and when to seek help

Play free now ▶