Japanese Numbers 1 to 100 (and Beyond), With Readings

How the Japanese number system works

Japanese numbers are built from ten core words, stacked in a regular pattern, so once you know 1 to 10 you can build almost everything else. Where English keeps inventing new words (eleven, twelve, twenty, thirty), Japanese mostly just glues the words you already know together. Eleven is literally ten-one. Twenty is two-ten. Thirty-five is three-ten-five. The structure is so consistent that, after a single afternoon with the first ten numbers, the path from there to 100 is more arithmetic than memorization.

There are only two kinds of wrinkles to learn, and both are small. The first is that a few numbers have more than one accepted reading, so 4, 7, and 9 each come in two flavors and you choose based on context and habit. The second is a handful of sound changes, where two sounds press together and one of them softens, the way English speakers say "handbag" closer to "hambag" without thinking about it. These changes cluster at predictable places (300, 600, 800, and a couple in the thousands), and once you have met them a few times they stop surprising you. The rest of this post walks through the system in order, with a table for each band of numbers and plain explanations between them.

Zero to ten

These ten words are the entire foundation, so it is worth learning them until they are automatic. Everything above ten is assembled from this list.

Notice that 4, 7, and 9 each list two readings. This is the one place in the first ten where you have a genuine choice, and it is the source of most early confusion. As a beginner, lean on yon (4), nana (7), and kyuu (9) as your everyday defaults. They are the safest, they are the most common in counting and in numbers read aloud, and they avoid a few awkward overlaps that the other readings can create. You will still meet shi, shichi, and ku in fixed expressions and in certain counters, so it is good to recognize them, but you do not have to reach for them yourself yet. Zero has two everyday forms too: rei and zero. The borrowed zero is what you will hear most often in daily life, for example when reading out phone numbers.

Eleven to ninety-nine

From eleven to ninety-nine, you build each number by stacking the words you already know, with no new vocabulary required. The pattern is consistent: a number above ten is read as tens-plus-ones, in that order. Eleven is 十一 (juuichi), literally ten-one. Twelve is 十二 (juuni), ten-two. To make the higher tens, you put a digit in front of ten: twenty is 二十 (nijuu), two-ten; thirty is 三十 (sanjuu), three-ten. Here are the round tens.

To say anything in between, you just add the ones digit to the end. Take thirty-five: start with thirty, 三十 (sanjuu), then add five, 五 (go), to get 三十五 (sanjuugo). Three-ten-five. Or take ninety-nine, the last number before the hundreds begin: ninety is 九十 (kyuujuu), and nine is 九 (kyuu), so ninety-nine is 九十九 (kyuujuukyuu). It reads exactly as it is built, nine-ten-nine, with no surprises. This is also where leaning on yon for 4 and nana for 7 pays off, because those readings keep the tens (yonjuu, nanajuu) clean and even. Practice a few random two-digit numbers out loud and you will feel how mechanical it is: pick a tens word, attach a ones word, done.

Hundreds, and the sound changes

The hundreds follow the same stacking logic, with the word 百 (hyaku) for one hundred, but three of them change their sound and those three trip up nearly everyone. The base pattern is to put a digit in front of hyaku: two hundred is 二百 (nihyaku), four hundred is 四百 (yonhyaku), five hundred is 五百 (gohyaku). But at 300, 600, and 800 the sounds press together and soften.

Here are the three changes, stated plainly so you can drill them on their own:

• 300 is さんびゃく (sanbyaku), not "sanhyaku". The hyaku softens to byaku.
• 600 is ろっぴゃく (roppyaku), not "rokuhyaku". The roku and hyaku compress into roppyaku.
• 800 is はっぴゃく (happyaku), not "hachihyaku". The hachi and hyaku compress into happyaku.

Everything else in the hundreds is regular, so these three are genuinely the whole list of exceptions to learn here. It helps to notice the pattern behind them: the trouble spots are the numbers built on three, six, and eight, which is the same family of digits that causes sound changes elsewhere in the language. If you commit just these three forms to memory, the hundreds stop being a problem.

Thousands, ten-thousands, and how Japanese groups large numbers

The thousands work like the hundreds, with the word 千 (sen) for one thousand and two more sound changes to watch, but the bigger idea is that Japanese groups large numbers by ten-thousand rather than by thousand. First the thousands themselves.

The two changes mirror what you saw in the hundreds: 3,000 is さんぜん (sanzen), where sen softens to zen, and 8,000 is はっせん (hassen), where hachi and sen compress into hassen. Again it is the three-family and the eight-family causing the wrinkle, which is a comforting bit of consistency once you notice it.

Now the part that genuinely disorients English speakers. In English, large numbers are grouped in threes, and we get a brand-new word every three zeros: thousand, million, billion. Japanese instead has a dedicated unit for ten-thousand, 万 (man), and counts large numbers in groups of four zeros. Ten-thousand is its own word, not "ten hundred-times-something".

Read those literal meanings slowly, because they are the crux of it. One hundred thousand is not its own word in Japanese; it is 十万 (juuman), ten of the ten-thousand unit. One million is 百万 (hyakuman), one hundred of the ten-thousand unit. So when you see a large figure written in digits, the trick is to stop grouping it the English way (in threes) and start grouping it in fours, counting up from the right. The four-zero block is where the unit name lives. This single shift is the most jarring thing about Japanese numbers for English speakers, and there is no shortcut: it just takes repetition until your eye learns to slice numbers into fours. Be patient with yourself here. Even fluent second-language speakers slow down for a moment when converting a big number out loud, and that is completely normal.

Reading numbers vs counting things

Reading a number aloud and counting objects are two different skills in Japanese, and this post has only covered the first. Everything above, from 一 (ichi) to 百万 (hyakuman), is what you use to read a price, a year, a phone number, a room number, or any quantity stated on its own. That is the abstract number system, and it is the foundation, which is why it is worth getting solid before anything else.

Counting actual things, though, adds a layer. When you count objects in Japanese, you attach a small word called a counter to the number, and which counter you use depends on what you are counting: flat things, long thin things, small animals, people, machines, and so on each take their own counter, and some of those combinations cause yet more sound changes. That is a whole topic of its own, and it is the natural next step once these base numbers feel comfortable. For now, hold the two ideas apart in your mind: this post is about reading the numbers, and counting things builds on top of it. When you are ready to add counters, the counters post picks up exactly where this one leaves off.

How to drill numbers so they stick

The fastest way to make numbers automatic is to read real ones out loud, every day, in small doses. Numbers are everywhere, which makes them easy to practice without setting aside special study time. If you want a fuller picture of how numbers connect to the rest of the language, the Japanese numbers topic hub gathers the related lessons in one place. Here is a calm routine that works.

• Read prices in Japanese. Whenever you see a price tag, a receipt, or a menu, quietly say the figure to yourself in Japanese. Prices push you straight into the hundreds and thousands, which is exactly the range that needs the most reps. If you would rather drill from a focused list, the N5 number vocabulary set covers the core words at exam level.
• Read the clock. Glance at the time and say the digits in Japanese. It is a tiny, frequent rep that you can do dozens of times a day without thinking of it as study.
• Drill the three hundred-sound-changes on their own. Practice さんびゃく (sanbyaku), ろっぴゃく (roppyaku), and はっぴゃく (happyaku) as a separate little set, plus さんぜん (sanzen) and はっせん (hassen) for the thousands. These five forms are where most mistakes live, so giving them dedicated attention pays off out of proportion to their number.
• Practice slicing big numbers into fours. Take any large figure and group it from the right in blocks of four zeros, then read it using 万 (man). Doing this deliberately, slowly, is how the ten-thousand grouping stops feeling foreign.

Underneath all of this, let spaced repetition carry the long-term memory. Numbers fade fast if you meet them once and move on, but they hold well when you review them at growing intervals, just before you would have forgotten. That is the whole logic of a spaced repetition system: it schedules each item to come back at the right moment so you spend your effort only where it is needed. Inku is a calm iPhone flashcard app built on exactly that approach, with FSRS scheduling and bundled audio so you can hear each number as you review it, and it runs locally with no account and no ads. Whether you use an app or a paper deck, the principle is the same: short daily reps, honest review, and patience. The numbers will settle in faster than you expect.

Common questions

Related reading

• Japanese counters, and why you cannot just count to ten
• N5 number vocabulary
• the Japanese numbers topic hub
• JLPT N5 vocabulary list
• Practice kana with the free quiz

Try Inku, free for 7 days, or read the full roadmap.