HKUDS/VideoRAG

[KDD'2026] "VideoRAG: Chat with Your Videos"

61
/ 100
Established

Combines visual content, audio, and text through a graph-driven knowledge indexing approach to enable retrieval-augmented generation over videos of extreme length—supporting hundreds of hours on a single GPU. Integrates with multi-modal LLMs for natural language question-answering and provides a cross-platform desktop application (macOS, Windows, Linux) with drag-and-drop video upload, multi-video analysis, and research-grade evaluation on the LongerVideos benchmark dataset containing 134+ hours of diverse content.

2,782 stars. Actively maintained with 1 commit in the last 30 days.

No Package No Dependents
Maintenance 13 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 22 / 25

How are scores calculated?

Stars

2,782

Forks

387

Language

Python

License

Last pushed

Jan 11, 2026

Commits (30d)

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/rag/HKUDS/VideoRAG"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.