TwelveLabs Marengo 3.0 Goes GA in Amazon Bedrock for Video Semantic Search

AWS ML Blog · rss · 2026-09-11

TwelveLabs Marengo Embed 3.0 is now GA in Amazon Bedrock Knowledge Bases, jointly encoding video, audio, images, and text into a 512-dim vector space for natural-language media search — no more stitching transcription, frame extraction, and vector DB pipelines. Includes a full walkthrough (S3 ingestion, 4-second segmentation, console testing with timestamped results, Retrieve API integration), use cases across sports/media/security/education/retail, availability in us-east-1 and us-west-1, and Bedrock-standard invocation pricing.

Original post →

More from Infra

Infra channel →