Our corpus holds 33,233 graded entries across six languages, and the rungs of its A1 to C2 ladder are unevenly spaced. Spanish puts 3,490 of its 4,710 entries at A1, A2 and B1, drops to 200 at B2 and 200 at C1, then climbs back to 820 at C2. The other five languages have the same shape. These are counts of our current packs, including gaps in B2 and C1 coverage; they are not estimates of how many words a CEFR level requires. The per-language counts match almost exactly, on purpose: one graded list, carried into six languages. That sameness tells you something about the ladder and something about our choices.
What the corpus holds
The whole corpus is 33,233 entries in 617 shipping packs. Six learner languages are paired with English in both directions: Spanish, French, German, Italian, Portuguese with a Brazilian voice, and Polish. Together the six account for 31,998 entries: Spanish 4,710, French 4,708, German 5,956, Italian 5,956, Portuguese 5,960 and Polish 4,708. The English-only packs hold 770 entries across B1 to C2, including the free 495-entry B1 starter, and one pair without English, Polish to Italian at A2, adds the last 465.
Every entry carries a CEFR grade from A1 to C2 and one of twelve themes, from Food & Drink and Travel & Places through Work, Money & School to Grammar & Function Words. Both directions count. A Spanish A1 figure of 1,250 is 625 headwords for an English speaker learning Spanish plus 625 for the reverse pair, and the per-language statistics pages show both columns if you prefer the halved figure.
Entries per level, six languages side by side
The table counts entries as they ship today, both directions included. Read down a column for one language's ladder. Read across a row to see how little the languages differ.
| Level | Spanish | French | German | Italian | Portuguese | Polish |
|---|---|---|---|---|---|---|
| A1 | 1,250 | 1,248 | 2,496 | 2,496 | 2,500 | 1,248 |
| A2 | 1,250 | 1,250 | 1,250 | 1,250 | 1,250 | 1,250 |
| B1 | 990 | 990 | 990 | 990 | 990 | 990 |
| B2 | 200 | 200 | 200 | 200 | 200 | 200 |
| C1 | 200 | 200 | 200 | 200 | 200 | 200 |
| C2 | 820 | 820 | 820 | 820 | 820 | 820 |
| Total | 4,710 | 4,708 | 5,956 | 5,956 | 5,960 | 4,708 |
Three things stand out. B1 is 990 in every language, B2 and C1 are 200 in every language, and C2 is 820 in every language. Only one place breaks the symmetry: the A1 row. And the bottom three rungs together hold between 74 percent of a language (Spanish, 3,490 of 4,710) and 80 percent (German, 4,736 of 5,956), which is the part of the ladder daily conversation runs on. The percentages are computed from the table.
Why A1 is twice as big in German, Italian and Portuguese
German and Italian list 2,496 entries at A1 and Portuguese 2,500, against 1,250 for Spanish and 1,248 for French and Polish. The beginner vocabulary is the same size in all six. The packaging differs. For German, Italian and Portuguese we ship A1 twice: once as a single starter set and once more split across the themed sets, so the same beginner words are carried by two packs at the same level and are counted in both. Spanish, French and Polish ship A1 once.
That is a curation decision, and we would rather show the double count than quietly halve it, so the per-language statistics pages disclose it and list entries next to distinct phrases. It is also the whole reason the headline totals for German, Italian and Portuguese run about 1,250 ahead of Spanish and French: 5,956 against 4,710. Strip the second A1 and German, Italian and Portuguese land on the same numbers as Spanish and French, 4,708 and 4,710.
The plateau at B2 and C1
Every language drops to 200 entries at B2 and 200 at C1, which is 100 headwords per direction at each level and the smallest tier on the ladder. These tiers currently cover only five of the twelve themes. That is a coverage gap in the shipping packs, not a claim that these levels need only 100 words.
For a learner, phrases, collocations and reading remain useful alongside a vocabulary pack: they show how familiar words combine and how the same verb takes on different meanings. The current B2 and C1 packs are a small selection to practise with; they cannot stand in for the full vocabulary needed at those levels.
B2 and C1 need broader coverage. The counts here describe what is available now, and will change when more packs ship.
Why C2 is 820 in every language
C2 is the surprise in the table. After two rungs of 200 the ladder widens again to 820 entries, more than four times the size of B2, and it is 820 in all six languages. This comparison reflects the current pack coverage, especially the smaller B2 and C1 sets. Newspapers, contracts, reports and novels use vocabulary from many levels; they do not belong exclusively to C2.
It is 820 everywhere because the C2 list was graded once and carried into each language whole. The current C2 set contains 410 entries per direction in each language. The same is true of B1, B2 and C1, which is why those rows are flat across the table. The ladder is a property of the list before it is a property of any language.
What an entry is
An entry is a word or a short headword, its translation, and one example sentence that shows it in context. Each is graded A1 to C2 and tagged with one of the twelve themes. The headword list comes from the New General Service List, version 1.2, which Browne, Culligan and Phillips publish under a CC BY-SA 4.0 licence. The grading, the theme curation, the statistics on this page and the audio are ours.
The audio is text-to-speech. What plays when the phone is plugged in is the target word, a pause, the translation, a pause, and the target word again. The example sentence is shown on the screen and is never spoken during a plug-in; its own audio can be tapped on the phrase page where available. That matters for reading this table: an entry is one word you will hear with one sentence you will read, so the 990 at B1 is 495 words heard and 495 sentences read in your direction of the pair.
What these numbers do not tell you
The table is a count of our list. It is a curated corpus with a grade on every entry, and the size of each level reflects the packs currently available. Nothing in the table measures the language itself. Spanish has far more than 4,710 words, and how many of them a learner needs is a different question, one the per-language word-count pages take up with the research bands.
Two more limits. The counts are entries, so where A1 ships twice the same word is counted twice, and the statistics pages carry the distinct figures. And the grades are ours: a word we placed at B1 might sit at A2 in a coursebook and at B2 in an exam list, because the CEFR describes what a learner can do, and every publisher maps vocabulary onto it by hand. Read the table as the shape of one graded list, built one way, and it holds. Treat it as the size of six languages and it will mislead you.
How to use the ladder
Start where you are. Look through a pack's examples and choose a level with useful words you do not yet know. Ordering food or reading a newspaper with a dictionary is not enough on its own to place you at a particular CEFR level. The app plays one word every time the phone is plugged in, and the pick is weighted towards the words you have gone longest without hearing, with a word never heard counting as about twenty days overdue, so a fresh level fills your first plug-ins with new words and settles into rotation later. One word per charge explains the rotation.
Then use the pages behind the table. Corpus statistics has the whole corpus by language and level, and each language has its own page with the per-direction counts, the themes and the entries-against-distinct disclosure: Spanish, French, German, Italian, Portuguese and Polish. The methodology page covers where the list comes from and how it is graded. Pick the rung and plug in.
Common questions
How many words do you need for B2?
Our current B2 packs contain 200 entries per language, 100 per direction, across five of the twelve themes. This is limited pack coverage, not the vocabulary requirement for B2. The per-language word-count pages discuss broader estimates and their limits.
Is A1 the same size in every language?
Almost. Spanish has 1,250 A1 entries, French and Polish 1,248. German and Italian show 2,496 and Portuguese 2,500 because those three languages ship A1 twice, once as a single starter set and once split across themed sets, and both are counted. The beginner vocabulary itself is the same size in all six.
How many words are in the Vocabcord corpus?
33,233 entries across 617 packs. Six learner languages paired with English in both directions account for 31,998 of them (Spanish 4,710, French 4,708, German 5,956, Italian 5,956, Portuguese 5,960, Polish 4,708); the English-only packs hold 770 entries across B1 to C2, including the free 495-entry B1 starter, and a Polish-to-Italian A2 pack holds the last 465. Every entry is graded A1 to C2 and themed, with one example sentence and text-to-speech audio.
Why is C2 bigger than B2 and C1?
The current C2 packs contain 820 entries per language, while B2 and C1 have 200 each and cover fewer themes. Those counts describe the available packs, not the relative vocabulary requirements of the levels. The same underlying sets are carried into all six languages.
