Hive

API

Content moderation and AI-generated media detection API

Price: Free credits, then pay per use
Access: API key

About

Hive runs pretrained models that score images, video, text and audio for nudity, violence, hate, spam and AI-generated or deepfake content. Self-serve V3 keys are capped at 100 requests a day per model; higher volume needs an enterprise project set up with Hive.

What you can do with it

Score user-uploaded images and video for nudity, weapons, drugs and gore
Flag hate, bullying, sexual content and spam in chat messages and comments
Detect AI-written text, AI-generated images, deepfakes and synthetic audio

Get started

Try the models in Hive's Playgrounds, then request a moderation project
Copy the project's key from Integration & API Keys
Send an image URL or a text string to the sync endpoint

Example

# Image or video by URL (visual moderation project key)
curl -X POST https://api.thehive.ai/api/v2/task/sync \
  -H "Authorization: Token $HIVE_API_KEY" \
  -F "url=http://hive-public.s3.amazonaws.com/demo_request/gun1.jpg"

# Text (text moderation project key)
curl -X POST https://api.thehive.ai/api/v2/task/sync \
  -H "Authorization: Token $HIVE_TEXT_API_KEY" \
  -F "text_data=Hi there, this is a test task"

Details

Hosting: Hosted service
Available in: Worldwide
MCP server: None