Amazon Web Services announced on October 9, 2026, that Amazon Bedrock now supports the TwelveLabs Pegasus 1.5 video-to-text model. The integration allows applications to generate structured, time-coded metadata from video libraries.
Capabilities and Features
Amazon Bedrock now supports TwelveLabs Pegasus 1.5, a video-to-text model that generates structured, time-coded metadata from video. According to AWS, the model reads an entire video and labels what happens and when, turning footage into data that applications can query.
Users can define business requirements such as editorial segments, speaker changes, sports plays, or brand appearances. Pegasus 1.5 finds these moments across the video and returns them with timestamps. The model analyzes sights, sounds, and on-screen text in a single pass without requiring a pre-indexing or ingestion pipeline to be built first.
Availability and Access
Pegasus 1.5 is accessible through US and Global cross-Region inference profiles on Amazon Bedrock. Users can visit the Amazon Bedrock product page, access the model in the Amazon Bedrock console, or consult the Amazon Bedrock documentation for supported Regions, APIs, features, and pricing.
What to do
- Visit the Amazon Bedrock product page, access the model in the Amazon Bedrock console, or check the Amazon Bedrock documentation for supported Regions, APIs, features, and pricing.
Key facts and where they come from
- Amazon Bedrock now supports TwelveLabs Pegasus 1.5, a video-to-text model.
Amazon Bedrock now supports TwelveLabs Pegasus 1.5, a video-to-text model that generates structured, time-coded metadata from video.
- Pegasus 1.5 reads entire videos and labels events without pre-indexing.
Pegasus 1.5 reads an entire video and labels what happens and when, turning footage into data your applications can query.
- The model is accessible through US and Global cross-Region inference profiles.
Pegasus 1.5 is accessible through the US and Global cross-Region inference profiles.