Text tools
Score text for abuse, Hinglish and Tanglish included
Three classifiers each score the text from 0 to 1: a general multilingual one (toxic), one for Hindi typed in Roman letters (profanity, harassment, hate_speech, caste, communal, regional, misogyny) and one for Tamil typed in Roman letters (offensive). flagged says whether any score reached the threshold. Up to 4,000 characters; the first 256 tokens are read. Scores are a model's guess: use them to sort or to ask a person, not to decide alone.
post/v1/text/moderate
Request
- Base URL: https://api.dharun.dev
- URL: https://api.dharun.dev/v1/text/moderate
- Auth: none declared
Query parameters
thresholdstring
Score at which flagged becomes true, between 0 and 1. Default 0.5.
modelsstring
Comma-separated: general, hinglish, tanglish. Default all three.
Request body
Response
Success
Changes
No recorded changes to this operation across all 1 revision of this API.