The words worth learning first
The commonest 2,500 words of a language, in the order a corpus says you will meet them. The first 500 are free here, with a meaning and a part of speech on every row.
Choose a language
- SpanishEspañol
- GermanDeutsch
- FrenchFrançais
- ItalianItaliano
- PortuguesePortuguês
- EnglishEnglish
- TurkishTürkçe
- RussianРусский
- Japanese日本語
- Korean한국어
- Chinese中文
- Arabicالعربية
Questions
Where does the order come from?
From how often each word appears in a large corpus of everyday writing. Frequency is counted across all the forms of a word and attributed to the dictionary form, so a verb is ranked where a reader really meets it rather than being spread thinly over a dozen endings.
Why are these lists shorter than the ones with 5,000 or 10,000 words?
Because past about 2,500 a frequency rank stops meaning much. Rank 6,000 in a web corpus is not a harder word, it is a word that happens to be common in journalism. Established graded reader series top out near 2,500 headwords for the same reason.
Is every word in the list a real dictionary word?
Yes. Every headword is checked against the language's own Wiktionary entry, which is where the part of speech, the gender and the meaning come from. That check is what removes the proper nouns, single letters and corpus noise that raw frequency lists are full of.
Do I need the app?
Not for this page. The A1 list is free here and free in the app. A2 to B2 are part of Oqou Pro, and the app is also where a word can go from a list into a review queue, which is the part a page cannot do.
Has a native speaker checked these?
Not yet. The frequency order and the part of speech are mechanical, and the meanings come from a dictionary rather than being written by hand, but nothing here has been through native review. If you spot a meaning that is technically right and practically useless, we would like to hear about it.
