Pegasus 1.5 from TwelveLabs is now available in Amazon Bedrock
TwelveLabs Pegasus 1.5 is now available in Amazon Bedrock. The video language model turns long-form video into structured, timestamped metadata: define the moments that matter and get every matching segment back as JSON, with new video segmentation, multimodal prompting with reference images, and on-the-fly processing of up to two hours of video.
On October 9, 2026, TwelveLabs announced the release of Pegasus 1.5 on Amazon Bedrock. This video language model converts lengthy video content into structured, timestamped metadata. Users can specify moments of interest, and Pegasus 1.5 will return every matching segment as JSON, complete with start and end times and the requested fields.
Video content, which is usually unstructured and underutilized, can now be analyzed using Pegasus 1.5, which generates time-based metadata based on user-defined segment definitions. Unlike previous versions, Pegasus 1.5 also supports multimodal prompting, allowing users to combine text descriptions with visual references to locate specific elements within a video, such as products, logos, or people.
Additionally, Pegasus 1.5 can process videos up to two hours long in a single request, a significant improvement from the one-hour limit in earlier versions. The model can be accessed on Amazon Bedrock via the bedrock-runtime endpoint, requiring only an S3 URI or base64 string for input, with no pre-indexing needed. No pre-indexing is required when using Pegasus 1.5 on Amazon Bedrock.
Videos can be sent directly from Amazon S3, and results are returned in the response. This allows for direct analysis of videos stored in Amazon S3, simplifying the process and maintaining data within AWS boundaries.
Written by urgent.news from AWS Machine Learning's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
