Skip to content
AI Radar by Wagner Solution
CS
Menu
worth noting New models

Pegasus 1.5 is available in Amazon Bedrock for video analysis

clearly official source

AWS has made Pegasus 1.5 from TwelveLabs available in Amazon Bedrock. According to AWS, the model finds video segments based on specified criteria and returns structured metadata with timestamps. It is available in the us-west-2 region in on-demand mode.

On 9 October 2026, AWS announced the availability of Pegasus 1.5 from TwelveLabs in Amazon Bedrock. According to AWS, the model analyzes an entire video and returns segments matching user criteria as structured metadata with timestamps. It also supports prompts that include reference images.

The model offers two modes. The general mode responds to a text prompt and allows the response structure to be specified using JSON Schema. The time_based_metadata mode accepts definitions of the segments to find and returns their start times, end times, and requested fields. According to AWS, processing requires no prior indexing, and the data stays within the AWS data environment.

The model is available in the US West (Oregon) region, us-west-2, in on-demand mode under the identifier twelvelabs.pegasus-1-5-v1:0. Video can be supplied through an Amazon S3 URI or embedded in the request as base64, with a size limit of 25 MB in the latter case. Calls are supported through the InvokeModel and InvokeModelWithResponseStream APIs; the latter enables streaming of the response.

What changed

Why it matters

Timestamped results allow subsequent processing to target specific parts of a video. If the model correctly recognizes the specified criteria, it may reduce manual review of recordings; the structured output can be passed directly to another program.

Two audiences, two different impacts

What this means

01

For individuals

A developer can try analyzing a short video directly in a request and receive output following a custom JSON Schema without preparing an index.

What to do During your first test, check the stopReason field; a value of length means the response has reached the configured maximum and may be incomplete.
More practical updates →
02

For a business

Businesses can integrate searches for specific situations in video footage into their workflows. When assessing the use of internal recordings, a key consideration is the claim by AWS that the data stays within its data environment.

Processes
What to decide To track costs for individual applications, create an application inference profile, add tags, and use its ARN instead of the model identifier when making calls.
More business impacts →
Amazon Bedrock Amazon S3 Pegasus 1.5 TwelveLabs

Check the original

Event sources

clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.

1
AWS Machine Learning Blog primary source · first detected Pegasus 1.5 from TwelveLabs is now available in Amazon Bedrock