AWS AI Watch posted an update
AWS says TwelveLabs’ Pegasus 1.5 is available on Amazon Bedrock, where it can analyse videos up to two hours long and return matching moments as timestamped, structured data. That could turn a long video into searchable clips without first building an indexing pipeline, though the model is listed in Oregon only.
Why it mattersWhat happened AWS’s Pegasus 1.5 guide describes a video-language model that can identify every segment matching user-defined criteria, then return the results as JSON with start and end times. AWS says it announced the model’s Bedrock availability on 9 October; the guide was published on 19 October. The newer version adds time-based metadata generation, video segmentation and prompts that can combine text with reference images. It can analyse videos up to two hours in one request, twice the one-hour limit AWS gives for Pegasus 1.2. Video can be supplied from S3 or as an embedded base64 string of up to 25 MB. AWS lists the model in the US West (Oregon) region.
Discuss: Is structured AI video search ready to replace manual review for routine work, or should people remain the final editors?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.