Twelve Labs is a company that provides developers with a powerful video understanding infrastructure to help them build programs that can see, hear, and understand the world as we do.

Why work with us

Twelve Labs provides state-of-the-art video foundation models to organizations that have extensive video data. Our models enable natural language search across large video libraries and can generate detailed descriptions and chapters of video, with our APIs easily integrating into any workflow.

By using our models, customers are able to unlock use cases such as content repurposing, clip-based licensing, digital fan experiences, and enhancement of end user content discovery/recommendation. Here are demo videos of our semantic video search and video-to-text generation in action :)