The 100 Most Common Chinese Characters

By Ray Banks·July 18, 2026·Characters

The most common Chinese character is 的 (de), the possessive particle — it alone accounts for about 3.5% of all running text. Below is the full ranked top 100, with pinyin, core meanings, and exact occurrence counts, measured across the 658 graded stories in the Literate Chinese catalog (141,730 character occurrences, July 2026). These 100 characters cover 56.7% of everything you'll read in those stories — which is why they're the highest-leverage hundred things a beginner can learn.

Where does this list come from?

Every list of "most common Chinese characters" depends on what text you count. Ours is measured across the 658 graded stories in the Literate Chinese catalog (July 2026) — 141,730 character occurrences of learner-oriented narrative text, from absolute-beginner dialogues to adapted classics. That makes it unusually honest about one thing: this is the frequency distribution of what a learner actually reads, not of newspaper editorials.

Full transparency about the method's fingerprints: a story corpus over-represents dialogue and daily life, and proper names leave traces — 约 (yuē) and 翰 (hàn) sit just outside our top 100 mostly because a popular story series stars a character named 约翰 (Yuēhàn, "John"). Every corpus has quirks like this; most lists just don't tell you.

In the tables, the traditional form follows in parentheses where it differs from the simplified (说/說). Where a character has multiple readings, we give the reading that dominates in running text, with the alternatives noted.

Characters 1–25

Characters 26–50

Characters 51–75

Characters 76–100

How much text do these 100 characters cover?

More than half. Measured across the same 658-story corpus, the top 50 characters cover 43.7% of all running text, the top 100 cover 56.7%, the top 200 cover 71.6%, and the top 300 cover 80.1%. The curve is steep at the start and flattens hard — 的 alone appears more often than the 90th and 100th characters combined, several times over.

The catch: coverage isn't comprehension. Even at 56.7% coverage, no complete story in our catalog reaches the ~98% threshold that reading research associates with comfortable comprehension on the top 100 alone. The top 100 are the skeleton of every sentence; the remaining characters are what the sentences are about. For the full staged breakdown of what each character milestone unlocks — from 300 through 3,000 — see how many characters you need to read Chinese.

How does this list compare to general frequency lists like Jun Da's?

The standard reference is Jun Da's Modern Chinese character frequency list (2004), compiled from a very large corpus of online text spanning news and fiction, and ranking nearly 10,000 characters. At the very top, the lists agree almost perfectly — 的, 一, 是, 了, 不, 人 dominate any Chinese corpus ever counted.

Below the top 20, the corpora diverge in an instructive way. Jun Da's corpus leans heavily on news and formal prose, so characters like 国 (國, guó, "country") — ranked #11 there — 经 (經, jīng), 民 (mín), and 产 (產, chǎn) sit high. In our graded-story corpus, 国 doesn't even crack the top 150. Meanwhile our list boosts the characters of dialogue and daily life: 你 "you" (#18), 笑 "laugh" (#53), 吃 "eat" (#57), 妈 "mom" (#65), 饭 "meal" (#87), 爷 "grandpa" (#89). Neither list is wrong — they're measuring different Chinese. If your near-term goal is reading stories, conversations, and fiction (which is where nearly every learner starts), a story-corpus ranking is the better study order; if you mainly read news, weight a general list higher.

How should you actually learn these characters?

Not from this table. A ranked list is a great map and a terrible vehicle: most of the top 30 are grammatical particles and function words that only make sense inside sentences. Nobody learns 的, 了, 着, 得, and 地 from flashcard definitions — you learn them by seeing them do their jobs a few hundred times, which is exactly what happens in your first ten graded stories. (We've written before about why flashcards alone don't produce readers.)

The efficient loop looks like this: read stories pitched at your level, let frequency do the scheduling (the top 100 will each appear dozens of times whether you plan it or not), and reserve spaced-repetition flashcards for the characters that refuse to stick on their own. Because these characters are this frequent, a beginner story is effectively a flashcard deck someone hid inside a plot. If you want to see the top 100 in the wild today, try reading your first story — at beginner level, the text above is more than half of what's on the page.