TwelveLabs Pegasus 1.5 Lands on Amazon Bedrock – Your Videos Just Became Searchable

Amazon Bedrock now supports TwelveLabs Pegasus 1.5, a video-to-text model that automatically generates time-coded metadata from your footage, letting you query specific moments without knowing where to look. Available via US and Global cross-Region inference profiles, it analyzes video content in a single pass—no pre-indexing needed—and spots exactly what matters to your business, from speaker changes to brand appearances.