JLPT N3 Vocabulary List: ~1,800 New Words Beyond N4, by Domain

The JLPT N3 vocabulary list covers a cumulative recognition vocabulary of roughly 3,750 words. This page focuses on the ~1,800 words that are genuinely new at N3, organized by domain.¹²³ N3 is the bridge from beginner to intermediate. Its new layer is where Japanese vocabulary moves from concrete everyday words into abstract, social, and news-media register.

Overview

N3 vocabulary is cumulative. The widely cited "~3,750" figure is the total recognition vocabulary expected at N3. It already includes the ~1,500 words of the N4 base, which in turn contains the ~800 N5 words underneath.¹²³ The layer genuinely new at N3 is roughly ~1,800 words.

This article lists only that new layer. The N4 base it sits on is covered in the "JLPT N4 Vocabulary List," and the foundation under that is covered in the "JLPT N5 Vocabulary List"; neither set is re-listed here.

Both the ~3,750 cumulative figure and the ~1,800-new figure are the widely used unofficial community count, not an official tally. The Japan Foundation and JEES (Japan Educational Exchanges and Services) publish no "Test Content Specification" listing vocabulary, kanji, or grammar items. Their stated reasoning is that the goal of study is communication, not memorizing item lists.⁴ This has held since the five-level test (N1–N5) was adopted beginning with the 2010 sittings, so every N3 list in circulation is a community reconstruction.⁴

Every N3 word count here is an unofficial community figure

No official JLPT vocabulary list has existed since the 2010 test revision.⁴ The figures on this page come from durable community reconstructions (Tanos, jpdb, JLPTsensei). They are cross-checked against one another and presented as the de facto standard, not as a JEES count.

Exam mechanics, such as sections, scoring, and sitting dates, belong to "The JLPT Explained: Levels, Sections, and What Each Means," and general word-count and coverage theory belongs to "How Many Japanese Words Do You Need to Be Fluent?"; this page does not re-derive either.

How N3 builds on N4 and N5

Cumulative N3 (~3,750) is approximately the full N4 base plus roughly ~1,800 new words. That N4 base is about ~1,500 words and already contains the ~800 N5 words.¹²³ The new layer alone is larger than the entire cumulative N4 inventory.

That makes N3 the biggest single-level vocabulary jump in the JLPT series. It marks the shift from "beginner finishing line" to "intermediate," crossing decisively from concrete everyday vocabulary into abstract, social, and news-media register.²

For the upper-beginner set this page assumes, see the "JLPT N4 Vocabulary List." For the foundation under it, see the "JLPT N5 Vocabulary List." This page lists only what N3 adds, organized by domain.

The N3 hook: the Sino-Japanese (漢語) surge

N3 is the level where kango (漢語, Sino-Japanese on'yomi compounds) density jumps sharply. The new layer is dominated by abstract two-kanji compounds such as 社会 (society), 政治 (politics), 環境 (environment), 経済 (economy), 影響 (influence), and 関係 (relationship). Many also have する-verb forms. By contrast, N5 and N4 leaned more on native 和語 (wago).⁵²

There is a register mechanism behind this. Kango makes up roughly 60% of the words in modern Japanese dictionaries but only about 18–20% of words in ordinary speech. Its share rises in formal or literary contexts and in the expression of abstract or complex ideas.⁶ As N3 vocabulary moves into abstract, social, and news-media territory, the proportion of kango among the new words rises with it.

How these compounds are built and read is the subject of "Jukugo (熟語): How Kanji Combine to Form Japanese Words"; this section only names the pattern.

Why "~1,800 new" and not an exact number

The count is an unofficial community consensus, not a JEES list, and the sources disagree. Cumulative N3 totals run roughly ~3,500–3,750+ because each source reconstructs the post-2010 list differently. Coto cites 3,500–3,750, while Tanos cites ~3,750.³¹⁴ The new-layer delta is also a range: the Tanos exam-spec reconstruction yields about 1,804 new-at-N3 words, while jpdb's frequency-derived list reports 2,976 entries on a different, corpus-media basis.²⁷ This page uses "~1,800 new" as the round figure most aligned with the exam-spec reconstruction.

How to read this list

Each domain table below lists words new at N3 with four reference columns: the kanji form, its kana reading, optional romaji, and an English gloss. Several tables add a column showing whether the entry also functions as a する-verb.

Because this page shows only what is new at N3, common N4 and N5 carryovers are intentionally absent. The cumulative N3 inventory is the N4 list plus these words.² Many kango entries double as する-verbs, so the noun and its verb form are best learned together.⁵

What's new at N3: vocabulary by domain

The entries below are representative anchors confirmed on the new-at-N3 community lists, not the full ~1,800-word inventory.⁵ Each domain table can be extended from the same sources. The vocabulary-membership citation applies to the word, not to the example sentence. The example sentences are verbatim Tatoeba sentences cited by ID.⁸

Everyday life, leveled up (abstract & concrete nouns)

These are everyday-register nouns with a level of abstraction the N4 set lacked. A reader meets them in ordinary conversation, but they name situations, impressions, and traits rather than concrete objects.⁵

Kanji form	Reading	Romaji	Gloss
状況	じょうきょう	jōkyō	situation, circumstances
都合	つごう	tsugō	convenience, circumstances
内容	ないよう	naiyō	content, substance
印象	いんしょう	inshō	impression
性格	せいかく	seikaku	personality, character
努力	どりょく	doryoku	effort (also する-verb)
不足	ふそく	fusoku	shortage, insufficiency (also する-verb)
大部分	だいぶぶん	daibubun	the greater part, most

状況じょうきょうは最悪さいあくだ。⁸
"The situation is awful."

どこが都合つごうがいい？⁸
"Where is convenient for you?"

彼かれは性格せいかくがよい。⁸
"He has a good personality."

第一だいいち印象いんしょうは大切たいせつだ。⁸
"First impressions are important."

Work, society, and daily systems (社会・仕事)

The social and institutional register opens at N3. This includes words for meetings, pay, law, and the systems that organize daily life.⁵ Many are kango, and several double as する-verbs.

Kanji form	Reading	Romaji	Gloss	Also する-verb?
会議	かいぎ	kaigi	meeting, conference	n/a
給料	きゅうりょう	kyūryō	salary, pay	n/a
法律	ほうりつ	hōritsu	law	n/a
税金	ぜいきん	zeikin	tax	n/a
販売	はんばい	hanbai	sales, selling	yes (販売する)
議会	ぎかい	gikai	parliament, assembly	n/a
外交	がいこう	gaikō	diplomacy	n/a

会議かいぎは何時なんじから？⁸
"What time does the meeting start?"

給料きゅうりょうはいくら？⁸
"What is your salary?"

税金ぜいきんをお支払しはらいください。⁸
"Please pay the tax."

彼かれが販売部はんばいぶの責任者せきにんしゃだ。⁸
"He's the one in charge of the sales department."

News, media, and the wider world (ニュース・報道)

This is the register that makes simple news and non-fiction readable: government, economy, environment, incidents, and the language of increase and decrease.⁵ It is the N3 payoff because it unlocks easy-news and graded-reader immersion.

Kanji form	Reading	Romaji	Gloss	Also する-verb?
事件	じけん	jiken	incident, case, affair	n/a
政府	せいふ	seifu	government	n/a
経済	けいざい	keizai	economy, economics	n/a
環境	かんきょう	kankyō	environment	n/a
影響	えいきょう	eikyō	influence, effect	yes (影響する)
発表	はっぴょう	happyō	announcement	yes (発表する)
増加	ぞうか	zōka	increase	yes (増加する)
記事	きじ	kiji	(news) article	n/a

新あたらしい政府せいふが選挙せんきょされた。⁸
"A new government was elected."

彼かれは経済けいざいの専門家せんもんかだ。⁸
"He is an expert in economics."

人ひとは環境かんきょうの産物さんぶつである。⁸
"Man is a product of his environment."

犯罪はんざいが増加ぞうかしている。⁸
"Crime is on the increase."

Emotion, opinion, and description (感情・意見)

N3 expands emotion and opinion vocabulary well past the N4 basics. It adds words for being moved, satisfied, patient, and for agreeing or opposing.⁵ Several are な-adjectives that carry an evaluative judgment.

The な-adjectives stay visibly distinct from the い-adjectives: 立派 (splendid) and 確か (certain) in this set both take な before a noun.

Kanji form	Reading	Romaji	Gloss	Type
感動	かんどう	kandō	being moved, deep emotion	noun / する-verb
満足	まんぞく	manzoku	satisfaction	noun / な-adj / する-verb
我慢	がまん	gaman	patience, endurance	noun / する-verb
反対	はんたい	hantai	opposition, the opposite	noun / する-verb
賛成	さんせい	sansei	agreement, approval	noun / する-verb
立派	りっぱ	rippa	splendid, fine	な-adjective
確か	たしか	tashika	certain, sure	な-adjective

とても感動かんどうしたわ。⁸
"I was very moved."

我慢がまんできない。⁸
"I can't stand it."

反対はんたいです。⁸
"I'm against it."

賛成さんせいです。⁸
"I agree."

More adverbs and onomatopoeia (副詞・擬音擬態語)

The adverb layer expands sharply at N3. It adds the connective and stance adverbs that structure intermediate speech and writing: 実は (actually), 結局 (in the end), 例えば (for example), やはり (as expected), 一方 (on the other hand), 確かに (certainly).⁵

Kanji / kana form	Reading	Romaji	Gloss
実は	じつは	jitsu wa	actually, to tell the truth
結局	けっきょく	kekkyoku	in the end, after all
例えば	たとえば	tatoeba	for example
やはり / やっぱり	やはり	yahari	as expected, after all
一方	いっぽう	ippō	on the other hand
確かに	たしかに	tashika ni	certainly, indeed

実じつは、姉あねなんです。⁸
"Actually, she's my older sister."

私わたしは結局けっきょく失敗しっぱいした。⁸
"I failed after all."

例たとえば、これはペンです。⁸
"For example, this is a pen."

N3 also brings a recognition wave of onomatopoeia and mimetics (擬音語 and 擬態語). The mimetic system is large enough to study on its own. Its four classes and how to read them are covered in "Japanese Onomatopoeia: The Four Classes (giongo, gitaigo)," which is the place to drill them.

Sino-Japanese する-verbs and compound expressions (漢語＋する)

This bucket captures the kango surge as verbs. A two-kanji compound noun plus する yields a verb. At N3, this is the dominant new verb type: 説明する (to explain), 比較する (to compare), 影響する (to influence), 成功する (to succeed), 失敗する (to fail), 利用する (to use).⁵

Each of these nouns also stands alone, as in 説明 (explanation), 比較 (comparison), and 影響 (influence), so the noun and the する-verb are learned together. The compounding mechanics are the subject of "Jukugo (熟語): How Kanji Combine to Form Japanese Words."

Kanji form	Reading	Romaji	Gloss (as する-verb)
説明する	せつめいする	setsumei suru	to explain
比較する	ひかくする	hikaku suru	to compare
影響する	えいきょうする	eikyō suru	to influence, affect
成功する	せいこうする	seikō suru	to succeed
失敗する	しっぱいする	shippai suru	to fail
利用する	りようする	riyō suru	to use, utilize
報告する	ほうこくする	hōkoku suru	to report
努力する	どりょくする	doryoku suru	to make an effort

説明せつめいさせて。⁸
"Let me explain."

君きみの答こたえを先生せんせいのと比較ひかくせよ。⁸
"Compare your answers with the teacher's."

これも温暖化おんだんかの影響えいきょうか？⁸
"Is this another effect of global warming?"

報告書ほうこくしょは提出ていしゅつしたの？⁸
"Have you turned in your report?"

N3 kanji coverage

The kanji newly tied to the N3 tier number roughly ~370 characters. The cumulative N3 kanji requirement is commonly cited at ~650 "including kanji from N4 & N5."⁹¹ Both are hedged, unofficial figures. The JLPT-to-jōyō mapping is approximate, not official.

These characters extend the N4 set into grade-3, grade-4, and grade-5 jōyō (教育漢字, school-taught kanji) territory. To drill the characters behind these words, see "Grade 3 Jōyō Kanji (小3)," "Grade 4 Jōyō Kanji (小4)," and "Grade 5 Jōyō Kanji (小5)."

Because the kango surge means most new N3 words are on'yomi compounds, kanji and vocabulary now reinforce each other tightly. Each new character unlocks several compounds, and each compound drills its characters.⁶

How to actually learn the new N3 words

The general method for acquiring vocabulary by level is covered in "How to Learn Japanese Vocabulary: A Strategy by Level." What is specific to N3 is its pacing and density.

The new ~1,800 words are the series' largest single jump, so budget time accordingly.² As a worked estimate, ~15–20 new cards per day clears the new layer in roughly 90–120 days, about 3–4 months, on top of a solid N4 base.

Pace the kango surge by co-studying kanji and vocabulary

The kango density at N3 makes kanji-and-vocabulary co-study unusually efficient. The same on'yomi recurs across many compounds, so each new character you learn unlocks several words at once.⁶ Drill the kanji and the vocabulary together rather than as separate piles.

Two facts shape how much recall you actually need. The JLPT measures recognition through reading and listening, not production, so passive recognition of these words is the exam bar.⁴ N3 is also cumulative, so the exam still draws on all ~1,500 N4 words and the ~800 N5 base under them. The new ~1,800 are an addition, not a replacement.¹³

Recommended deck and list sources

The de facto standard sources for the new-at-N3 inventory are durable community reference lists, not an official document.⁴ The figures on this page rest on three:

• Jonathan Waller's Tanos N3 list, reconstructed from the pre-2010 test specification. Its new-at-N3 layer is rendered as ~1,804 words, with ~3,750 cumulative vocabulary and ~650 kanji.¹²
• jpdb's frequency-ordered N3 vocabulary list, which reports 2,976 entries on a corpus-media reconstruction.⁷
• JLPTsensei's N3 vocabulary and kanji references, with per-entry readings and glosses and a stated 370 N3-specific kanji.⁵⁹

The well-known commercial study series Tango N3 (JLPT Tango), Nihongo So-matome, and Shin Kanzen Master are widely used and reliable. Because they are paywalled print products, they are cited here as resources rather than as the basis for any word count.

Amenokori app callout

Drill the N3 delta with FSRS-scheduled, level-mapped decks

N3 is where the list lengthens sharply. To keep it from feeling like a wall, J-Compass recommends Amenokori: its N3 deck covers the delta directly, with FSRS spacing the larger volume so the daily queue stays steady rather than spiking.¹⁰¹¹ For where it fits among other tools, see "Choosing Your First Japanese Resources: Free vs. Paid" and "How to Learn Japanese Vocabulary: A Strategy by Level".

Good to know

Treating ~3,750 as words to learn fresh for N3

The ~3,750 figure is the cumulative total. It already includes the ~1,500 N4 words and the ~800 N5 base under those; only ~1,800 are new at N3.¹³ Budgeting study for 3,750 fresh words double-counts the N4 and N5 base you already hold.

Chasing a single "correct" N3 count

The new-at-N3 layer is reconstructed differently by each list. Tanos gives ~1,804 new words, jpdb gives 2,976 on a frequency basis, and cumulative totals run ~3,500–3,750+.²⁷³¹ None of these is canonical. They are unofficial post-2010 reconstructions, so chasing one exact number is wasted effort.⁴

Neglecting the N4 and N5 base while chasing the new layer

N3 is cumulative, so the exam still draws on all the lower-level words.¹³ The new ~1,800 are an addition, not a replacement, and the base must stay sharp.

The kango surge makes kanji study pay compounding interest

Kango is about 60% of dictionary vocabulary and concentrates in abstract and formal register.⁶ Since most new N3 words are on'yomi compounds, the same on'yomi recurs across many of them. Each new kanji learned can unlock several words at once.

Many N3 nouns double as する-verbs

Learn the verb form alongside the noun, as with the pairs below. The 漢語＋する pattern is the dominant new verb type at N3.⁵

説明せつめいします。⁸
"I will explain."

Recognition, not production, is the exam bar

The JLPT tests reading and listening, so passive recognition of the new N3 words is enough to pass.⁴ Productive recall means being able to use the words in speech and writing. It is a separate and higher goal.

News and media register unlocks easy-news and graded readers

The 事件 (incident), 政府 (government), 経済 (economy), 環境 (environment), 影響 (influence), 報道 (news reporting) layer is what makes simple-news and graded-reader immersion possible.⁵ Pairing the N3 list with that immersion is the most efficient way to lock the new vocabulary in.

See also

• JLPT Vocabulary by Level: How Many Words for N5 to N1
• JLPT N2 Vocabulary List: ~1,750 New Words Beyond N3, by Register
• JLPT N1 Vocabulary List: ~10,000 Words and Why You Read Instead of Drill
• Wago, Kango, Gairaigo, Konshugo: The Four Vocabulary Strata of Japanese
• On'yomi vs. Kun'yomi: The Two-Reading System Behind Every Kanji
• How Many Kanji Do You Need? A Realistic Count
• Passive vs. Active Vocabulary in Japanese: The Two-Speed Problem
• Suru-Verbs (する-Verbs): How する Turns Nouns Into Verbs
• Japanese Adjectives Overview: The Two Classes (い-形容詞 vs な-形容詞)

References

Footnotes

1. Jonathan Waller. "JLPT Level N3 Resources." tanos.co.uk. http://www.tanos.co.uk/jlpt/jlpt3/ . Community N3 resource set reconstructed by mapping the pre-2010 Test Content Specification to the new N-level structure, with frequency in the Tanaka corpus used to sort items whose N-level was unclear. The page frames the cumulative N3 requirement as "~3,750" vocabulary and "~650" kanji, and states explicitly: "There's no official JLPT N3 vocabulary, kanji or grammar list." Downloadable N3 list: http://www.tanos.co.uk/jlpt/jlpt3/vocab/VocabList.N3.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

2. Tanos N3 vocabulary list (new-at-N3 layer), as compiled and distributed by Jonathan Waller, tanos.co.uk. http://www.tanos.co.uk/jlpt/jlpt3/vocab/VocabList.N3.pdf . The Tanos per-level vocabulary lists are NOT cumulative: each level contains only the words newly required at that level, not the lower-level carryover. A widely-mirrored rendering of this list (Unseen Japan study PDF) titles it "JLPT N3 Vocabulary (1804 words)," giving the new-at-N3 layer as 1,804 items. https://media.unseen-japan.com/wp-content/uploads/2021/07/JLPT-N3-Vocabulary-.pdf ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰

3. Coto Academy. "JLPT N3 Vocabulary List." https://cotoacademy.com/jlpt-n3-vocabulary/ . States "JLPT N3 vocabulary includes about 3,500 to 3,750 words. This typically includes vocabulary you have learned in JLPT N5 and JLPT N4," and that its own list "includes only new vocabulary introduced at the JLPT N3 level. It does not cover words already studied in JLPT N4 and N5." (limitation: language-school blog; used only for the cumulative-figure band and the cumulative-not-new framing, both corroborated by ¹.) ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

4. Japan Foundation & Japan Educational Exchanges and Services (JEES). "Is there a 'Test Content Specification'?" JLPT FAQ. https://www.jlpt.jp/e/faq/ . Official statement that the test publishes no list of vocabulary, kanji, or grammar items. Direct quote: "We decided that publishing 'Test Content Specifications' containing a list of vocabulary, kanji and grammar items was not necessarily appropriate." In their place the JLPT offers a summary of competencies per level, the test composition, and sample questions. The same FAQ states the five-level test (N1–N5) and current passing standards were adopted "beginning with July test in 2010 (December test in 2010 with N4 and N5)." ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸

5. JLPTsensei. "JLPT N3 Vocabulary List." https://jlptsensei.com/jlpt-n3-vocabulary-list/ . Paginated N3 word reference with readings and English glosses; itemizes only the words newly tested at N3 (does not re-list N4/N5 carryover). Used here to confirm N3 membership of representative entries across the domain buckets. (limitation: language-learning site; used only for word-membership confirmation, corroborated by the durable Tanos reconstruction ¹².) ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹²

6. Shibatani, Masayoshi. The Languages of Japan. Cambridge University Press, 1990. Establishes that roughly 60% of the words in modern Japanese dictionaries are kango (Sino-Japanese on'yomi vocabulary), while only about 18–20% of words in ordinary speech are kango, and that kango usage rises in formal or literary contexts and in the expression of abstract or complex ideas. (Proportions as summarized in the standard reference literature; the speech-frequency figure derives from a NINJAL/Kokuritsu Kokugo Kenkyūjo word-count survey.) ↩ ↩² ↩³ ↩⁴

7. jpdb. "Japanese Language Proficiency Test – N3 – Vocabulary list." https://jpdb.io/vocabulary-list/924/japanese-language-proficiency-test/3/n3/vocabulary-list . Frequency-ordered N3 deck/label; the N3 vocabulary list header reports 2,976 entries. jpdb's N3 set is built from a different (corpus-frequency, media-derived) reconstruction than Tanos, which is why its count differs. ↩ ↩² ↩³

8. Tatoeba Project. Tatoeba: Collection of sentences and translations. https://tatoeba.org . Open, community-verified bilingual sentence corpus; individual sentences are cited by numeric ID. Each example below pulled verbatim via the Tatoeba API with its English translation. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷ ↩⁸ ↩⁹ ↩¹⁰ ↩¹¹ ↩¹² ↩¹³ ↩¹⁴ ↩¹⁵ ↩¹⁶ ↩¹⁷ ↩¹⁸ ↩¹⁹ ↩²⁰ ↩²¹ ↩²² ↩²³ ↩²⁴ ↩²⁵

9. JLPTsensei. "JLPT N3 Kanji List." https://jlptsensei.com/jlpt-n3-kanji-list/ . States "370 kanji are specific to N3 level" and that "You must know about 650 kanji in total in order to pass the JLPT N3, including kanji from N4 & N5." ↩ ↩²

10. Amenokori. Product landing page. https://amenokori.com . Mobile Japanese-learning app "Built around the FSRS algorithm so you study less and remember more," offering "Comprehensive SRS decks for vocabulary, grammar, and kanji" spanning "N5 (beginner) through N1 (advanced)" with "separate decks by level." Headline figures: "10K+ Words & grammar," "150K+ Quiz questions," and a 2,136-character jōyō kanji list described as "Every regular-use kanji, sorted by frequency" with on'yomi, kun'yomi, and meanings. ↩

11. Amenokori. Mobile-app page. https://amenokori.com/mobile-app/ . States the app is available on the App Store (iOS) and Google Play (Android), describes its scheduling as the same algorithm used in current Anki and calibrated to the user's memory, covers "N5 (beginner) through N1 (advanced)" with "separate decks by level" (plus an extended N1 vocabulary deck), and bundles "Vocabulary, grammar nuance, and kanji in a single app" with "meanings, registers, usage notes" and seven quiz types. ↩