modbender Dreamer Corpus
Access the COCA corpus API at https://api.dr.eamer.dev.
COCA contains 1+ billion words of contemporary American English from spoken, fiction, magazine, newspaper, and academic sources.
Authentication
export DREAMER_API_KEY=your_key_here
Endpoints
Word Search / Concordance
GET https://api.dr.eamer.dev/v1/corpus/search?word=serendipity&limit=20
Returns KWIC (keyword-in-context) examples showing the word in actual usage.
Collocations
GET https://api.dr.eamer.dev/v1/corpus/collocations?word=run&pos=verb&limit=20
Returns words that statistically co-occur with the target word (MI score, frequency).
Frequency
GET https://api.dr.eamer.dev/v1/corpus/frequency?word=algorithm&genre=academic
Returns frequency per million words, with optional genre filter: spoken, fiction, magazine, newspaper, academic.
When to Use
- Checking how formal or common a word is in real American English
- Finding natural collocations for writing assistance
- Linguistic research on word usage patterns
- Historical frequency trends across decades
Don't Use When
- You need non-English corpora
- You need corpora other than contemporary American English (COCA is 1990-present)