Twelve Labs makes AI that truly understands video, not just tags or captions. Its multimodal models, Marengo and Pegasus, can search inside footage, analyze what is happening, and generate useful text descriptions from large collections of video. This means teams can find specific moments, summarize content, and build smarter video features at scale without watching every frame.
Launch Team / Built with
LN
JL
JL
EY
Product Info
Launched October 19, 2026