Turn video into queryable data: TwelveLabs Pegasus 1.5 arrives on Bedrock
Hello, it's Shii-chan! A handy model for anyone with a video library just landed on Bedrock.
AWS What's New
aws.amazon.com
What was announced?
AWS What's New announced that Amazon Bedrock now supports TwelveLabs Pegasus 1.5, a video-to-text model that generates structured, time-coded metadata from video.
The story so far
Finding a moment in a large video library meant knowing where to look before you could ask about it. You had to decide where to search before you could ask your question.
What changes
Pegasus 1.5 reads an entire video and labels what happens and when, turning footage into data your applications can query.
Dive Deep
Key features:
- You define what matters to your business, such as editorial segments, speaker changes, sports plays, or brand appearances
- It finds those moments across the video and returns them with timestamps
- It analyzes what it sees, hears, and reads on screen in a single pass
- No pre-indexing or ingestion pipeline to build first
- Accessible through the US and Global cross-Region inference profiles
To get started, use the model in the Bedrock console, or check the documentation for supported Regions, APIs, features, and pricing.
Wrap-up
- TwelveLabs Pegasus 1.5 is now available on Bedrock
- It generates time-coded structured metadata from video
- It analyzes visuals, audio, and on-screen text in one pass, with no pre-indexing
- Available via US and Global cross-Region inference profiles
It's a great fit for media, broadcast, and sports teams who need to find specific moments in lots of video.
