Jockey
Video intelligence agent searching and parsing large media libraries.
Quick Keypoints
- Orchestrates natural language searches across raw video assets.
- Leverages Marengo and Pegasus foundation models for video context.
- Extracts structured machine-readable metadata and clip locations.
- Integrates with assistants via Model Context Protocol support.
What is Jockey?
Video intelligence agent searching and parsing large media libraries.
Jockey is Twelve Labs' agentic video intelligence preview platform. Powered by Twelve Labs' Marengo and Pegasus video foundation models, Jockey processes raw media files to support natural language queries, trace actions across shots, and output structured metadata files.
Who Needs Jockey?
Video editors, media archiving teams, and developers building semantic media applications.
Primary Use Cases
- Streamlining team workflows and automations.
- Summarizing and organizing complex content assets.
- Enhancing day-to-day productivity using smart AI helpers.
Important Features
- Media Search: Retrieve clip intervals based on conceptual commands without tags.
- Structured Extraction: Automatically map scenes, timeline transcriptions, and keywords into clean JSON data.
- MCP Connectors: Allow external LLMs to query media database states through standard endpoints.
Current Updates About Jockey
- Jockey by Twelve Labs has opened its developer API research preview program.
- Current Version: Jockey Preview v1.0
Alternatives to Jockey
If you want to check similar software, these alternative tools offer comparative features:
Editorial Rating
4.8 / 5.0
Pricing Plans
| Plan | Price |
|---|---|
| Research PreviewAccess to Jockey developer dashboard, limited API credits, community Slack support | $0 |