How it works
The checker splits your text into words and looks up each one in a list of the 20 000 most frequent Swedish word forms, ranked from jag (number 1) downwards. It then shows what share of your text falls within the top 1 000, 2 000, 5 000, 10 000 and 20 000. Words outside the band you choose are marked in the text; click one to look it up in the Lingolib dictionary (not every word has a page yet).
Capitalised words in the middle of a sentence that are not in the list, such as Uppsala or Astrid, are treated as names and left out of the figures. Everything runs in your browser; your text never leaves your device.
Word forms, not dictionary words
The list counts word forms, so bil, bilen, bilar and bilarna each have their own rank. That matters in Swedish, where a noun has up to eight forms and a verb even more. A form such as bilarnas can rank as rare even though bil is one of the first words you learn. If a word you know is marked, that is probably why.
How to read the numbers
| Known words | What reading feels like |
|---|---|
| 98% or more | Comfortable. You can read for pleasure and guess the odd new word from context. |
| 95% | Hard but possible. About one unknown word in every line or two; you will reach for a dictionary. |
| 90% | Frustrating. One unknown word in every ten; you will lose the thread without help. |
| Under 90% | Study material rather than reading: work through it with a glossary or a teacher. |
The percentages here are shares of the text within the top bands, which is a proxy for how many words you know. If you know roughly the 2 000 most common forms and the checker shows 80% in the top 2 000, expect about one word in five to be new.
What the levels roughly correspond to
- A1–A2: almost everything is everyday vocabulary: easy readers, simple messages, children's books.
- A2–B1: mostly common words with some less frequent ones: graded readers, simple news (for example 8 Sidor).
- B1–B2: a fair share of less common words: popular fiction, blogs, regular news.
- B2–C1: rich vocabulary: newspaper features, opinion pieces, non-fiction.
- C1+: many rare or specialised words: academic, legal, technical or literary texts.
Vocabulary is only part of the difficulty. Long sentences, subordinate clauses and unfamiliar topics make a text harder, and a familiar topic makes it much easier. To check your grammar as well, take the Swedish level test.
Using the checker to learn
- Pick texts in your zone. Aim for texts where 95% or more of the words are ones you know; that is where you learn the most with the least frustration.
- Pre-learn the rare words. Look through the rarest words list before you read, and learn the ones that repeat.
- Build the base first. The top 2 000 forms carry most of any text. The most common Swedish words list is a good place to start.
Source and licence
Word frequencies: FrequencyWords by Hermit Dave, based on the OpenSubtitles 2018 corpus, licensed under CC BY-SA 4.0. The derived list used here (the top 20 000 forms, lowercased, without numbers and symbols) is shared under the same licence.
Frequently asked questions
How many Swedish words do I need to know to read a book?
Research on reading in a foreign language suggests you need to know about 95% of the running words to follow a text, and around 98% to read comfortably without a dictionary. For everyday fiction that usually means knowing several thousand word families. Paste a page into the checker to see where a particular book sits.
How many words are enough for everyday Swedish conversation?
The most common 1 000 word forms make up roughly three quarters or more of everyday spoken Swedish, and 2 000 to 3 000 forms cover most of it. That is why the first months of learning should focus on the most frequent words.
Is the CEFR estimate reliable?
It is a rough guide to vocabulary only. It ignores grammar, sentence length and topic, and the word list comes from film and TV subtitles, so it favours conversational Swedish: nature, politics or technical words can look rarer than they feel. Use it to compare texts with each other, not as an official grading.
Where does the word list come from?
From Hermit Dave's FrequencyWords project, which counted every word in the Swedish part of the OpenSubtitles corpus of film and TV subtitles. It is released under the CC BY-SA 4.0 licence. This tool uses the 20 000 most frequent word forms.
Is my text sent anywhere?
No. The word list is downloaded once and the analysis runs entirely in your browser. Nothing you paste is uploaded or stored.