Name: Korean Learner Dictionary API - KRDict JSON
Brand: 193market
SKU: api-193market-korean-learners-dictionary
Availability: InStock
Author: API Market

## The official Korean learner's dictionary, as a fast JSON API

Korean Learner's Dictionary API serves the *Korean Basic Dictionary* (한국어기초사전) published by
the National Institute of Korean Language, converted from its official XML release into clean,
paginated JSON. It is the dictionary that Korean language textbooks and government learning
programmes are built on, and this is the shortest path to putting it inside your own product.

If you are building a Korean learning app, a flashcard tool, a reading assistant, a chatbot or
an agent that has to explain a Korean word, you need three things that a general translation API
will not give you: graded learner levels, real example sentences, and definitions written for
learners rather than for native speakers. This dataset has all three, and this API makes them
queryable in one call.

### What is inside

- **56,555 headwords** — words, idioms, proverbs and grammar expressions.
- **76,833 senses**, each with a Korean definition written for learners.
- **594,578 examples**: phrases, full sentences and short dialogues.
- **Translations in 11 languages** — English, Chinese, Japanese, French, Spanish, Arabic,
  Mongolian, Vietnamese, Thai, Indonesian and Russian, 826,000 translation records in total.
- **Learner levels**: beginner (2,545), intermediate (9,068) and advanced (37,024) entries, so
  you can build a graded curriculum instead of guessing at difficulty.
- **Hanja origin, pronunciation, part of speech and semantic category** for every entry.
- Built from the **2026-08 official release**, the most recent one published.

### Search that works in both directions

Type `사랑` and you get the Korean entry. Type `love`, `学校`, `たべる` or `manger` and you get the
Korean words that mean it. Reverse search across all 11 languages is the feature that usually
costs the most to build yourself, and it is the default behaviour here.

Results are ranked so that the words a learner actually needs come first: an exact headword match
outranks a sense match, a beginner word outranks an advanced one, and a short common word
outranks a rare compound. You can further narrow by `level` and by part of speech.

### Endpoints

| Method | Path | What it returns |
|---|---|---|
| GET | `/v1/search` | Search by `q` in Korean or any of the 11 languages. Filters: `lang`, `level`, `pos`. Paginated. |
| GET | `/v1/words/{word_id}` | The full entry: every sense, example and translation. |
| GET | `/v1/random` | A random word, optionally restricted by `level`. |
| GET | `/v1/word-of-the-day` | One word per UTC date, the same for every caller that day. |
| GET | `/v1/stats` | Dataset counts, so you can show your users what is covered. |

### A worked example

`GET /v1/search?q=love&lang=en&level=beginner` returns the beginner-level Korean words for
"love", each with its definition and example sentences, which is exactly one call for a flashcard
deck. `GET /v1/word-of-the-day?lang=en` gives every user of your app the same word on the same
day without you having to store or schedule anything. `GET /v1/words/34525` returns the complete
entry for 사랑, senses and all.

### Who this is for

- **Language-learning apps** that need graded vocabulary with real examples rather than a
  word list scraped from somewhere.
- **Flashcard and spaced-repetition tools** that want definition, example, level and
  pronunciation in a single response.
- **Reading assistants and browser extensions** that look up a word the reader tapped.
- **Chatbots and AI agents** answering "what does this Korean word mean". The JSON is flat and
  the fields are explicit, so a model can use it without a parsing step.

### Why not build it yourself

The source release is a set of XML files in a legacy encoding, with senses, examples and
translations split across separate trees that have to be re-joined by identifier. Getting the
raw files at all requires an application on the Korean-language dictionary site. After that you
still need the reverse-language index and the ranking logic, or a search for "love" returns
several hundred entries in no useful order. That work is already done here, and the subscription
covers hosting, search, ranking and uptime.

### Limits, stated plainly

- **Audio recordings and illustrations are not included.** The source licence excludes them, so
  this API returns text only and never links to those files.
- **This is a learner's dictionary, not a full native dictionary.** It covers the vocabulary
  taught to learners of Korean, which is what makes the level grading meaningful, but it is not
  an exhaustive record of the language.
- **The data is a published release, not a live feed.** It tracks the official release, currently
  2026-08, and updates when the Institute publishes a new one.

### Source and licence

Dictionary content © National Institute of Korean Language, *Korean Basic Dictionary*
(한국어기초사전), https://krdict.korean.go.kr, licensed under Creative Commons
Attribution-ShareAlike 2.0 Korea (CC BY-SA 2.0 KR). This API converts the official XML release
to JSON, and every response carries an `attribution` object naming the source, the licence and
the changes made.

Please note what that licence means for you: the dictionary content you receive stays under
CC BY-SA 2.0 KR. Your subscription pays for the hosting, the search, the ranking and the uptime,
not for exclusive rights to the underlying text. Attribution travels with the data, which is why
it is in every response rather than buried in these docs.

## Base URL

```
https://prod.api.market/api/v1/193market/korean-learners-dictionary
```

## Authentication

Every request needs your API.market key in the `x-api-market-key` header. You can find it in the API Keys section of your dashboard after subscribing. No other credential is required.

## Endpoint reference

| Method | Path | Parameters |
|---|---|---|
| GET | `/v1/search` | `q` and the filters described above |
| GET | `/v1/words/{word_id}` | `word_id` in the path |
| GET | `/v1/random` | none |
| GET | `/v1/word-of-the-day` | none |
| GET | `/v1/stats` | none |

### Examples

Search in Korean or in English:

```bash
curl -G 'https://prod.api.market/api/v1/193market/korean-learners-dictionary/v1/search' \
  --data-urlencode 'q=사랑' \
  -H 'x-api-market-key: YOUR_API_KEY'
```

Fetch one entry by id, as returned by search:

```bash
curl 'https://prod.api.market/api/v1/193market/korean-learners-dictionary/v1/words/34525' \
  -H 'x-api-market-key: YOUR_API_KEY'
```

A random entry, useful for flashcards and daily-word features:

```bash
curl 'https://prod.api.market/api/v1/193market/korean-learners-dictionary/v1/random' \
  -H 'x-api-market-key: YOUR_API_KEY'
```

Corpus size and available translation languages:

```bash
curl 'https://prod.api.market/api/v1/193market/korean-learners-dictionary/v1/stats' \
  -H 'x-api-market-key: YOUR_API_KEY'
```

## Status Codes

| Code | Meaning |
|---|---|
| 200 | Success, results returned |
| 400 | Invalid or missing parameter |
| 401 | Invalid or missing API key |
| 429 | Rate limit or plan quota reached |
| 5xx | Upstream error, retry the request |

## Contact Us

If you have any questions or need assistance, you can reach out to us through the embedded chat on the bottom right corner of the page or send an email to hello@magicapi.com.


JSON API for the official Korean Basic Dictionary: 56,555 headwords, 594,578 examples, translations in 11 languages, graded beginner to advanced.

Korean Learner Dictionary API - KRDict JSON

Free