A conversational video agent: understands your entire video and photo library and finds the right moments by a query in natural language.

Description

Jockey is a conversational video agent from Twelve Labs that understands your entire video and photo library and finds the right moments by a query in natural language — without tags or file names. It is available via API and MCP integrations in Claude and ChatGPT.

Key features

  • Video search in natural language
  • Understands the library without tags
  • API for developers and MCP integrations
  • Trend intelligence from YouTube, TikTok, Reels
  • Marengo, Pegasus and Search models