Pegasus 1.6

See the unseen. Know the unknowable.

Visit website

Video-understanding AI model that turns video and images into searchable, structured text for developers and media teams.

TwelveLabs homepage hero

TwelveLabs homepage hero

1 / 3

Overview

What is Pegasus 1.6?

See the unseen. Know the unknowable.

Pegasus 1.6 is TwelveLabs' newest video-understanding model. It converts video and still images into structured, searchable text — summaries, tags, chapters, timestamps, and custom-schema JSON — through a single API, understanding vision, audio, speech, and on-screen text natively.

New in 1.6: first-person (egocentric) video understanding for robotics and physical-AI training pipelines (turning teleoperation footage into labeled training data with no manual annotation), stronger entity recognition for people, characters, and objects, and native image analysis through the same API and prompts.

Pricing is freemium: a Free plan with 600 minutes of indexing and no credit card, pay-as-you-go Developer pricing (Analyze API from $1.75/hour of video), and custom Enterprise contracts. Available via web console/playground and REST API with Python and Node.js SDKs.

Platforms and languages

Pegasus 1.6 Availability

Platforms

WebApi

Languages

EnglishKoreanJapanese

Capabilities

Pegasus 1.6 Key Features

Egocentric video understanding

Understands wearable and robot-mounted first-person footage for robotics and physical-AI training pipelines.

Stronger entity recognition

More accurate identification and naming of people, characters, and objects in video.

Native image analysis

One API, one model, and one prompt handle both video and still images — no second image pipeline to maintain.

Structured text in one call

From up to 2 hours of video, return summaries, tags, chapters, and timestamped metadata in a JSON schema you define.

Native multimodal analysis

Understands vision, audio, speech, and on-screen text together, not just frame-by-frame captions.

Best for

Who uses Pegasus 1.6?

Developers

Build video search and analysis into apps via API

Robotics and physical AI teams

Turn egocentric and teleoperation footage into robot training data

Media and broadcasting teams

Catalog and search large video archives

Plans and access

Pegasus 1.6 Pricing

Freemium

Free 600 min; Developer pay-as-you-go (Analyze $1.75/hr video); Enterprise custom.

Free trial available

Common questions

Pegasus 1.6 FAQs

Do I need a credit card for the Free plan?

No. Signing up is enough to use the Free plan; no credit card is required.

How do the 600 free minutes work?

You can upload up to 600 minutes of video in total. Usage is counted cumulatively and is not restored if you delete uploaded indexes.

How can I access Pegasus 1.6?

Web console/playground plus a REST API with official Python and Node.js SDKs.

Can the model be deployed privately?

Yes, private or self-hosted deployment is possible — contact sales for details.

How is input video duration billed?

Billed duration is the video window you submit; for Segment, the number of segment definitions in your request also counts.

Reviews

See what the community thinks and share your experience.

—

Based on 0 ratings

Rating distribution

0
0
0
0
0

Leave a review

Sign in to rate

Community reviews

No reviews

No written reviews yet.

Explore by category

AI Development Infrastructure

View all AI Development Infrastructure websites

Explore by category

AI Research Analysis

View all AI Research Analysis websites
DecisionsApi preview
DecisionsApi

Explore decision models in one playground and API. Classify, score, and route text with structured answers and probability signals.

AI Research Analysis

Explore by category

AI Specialized Tools

View all AI Specialized Tools websites
MyToolster preview
MyToolster

MyToolster offers 50+ free browser-based web tools with no signup — AI detectors, background remover, audio/video to text, JSON, Base64, QR tools and more.

AI Specialized Tools
Cubicle preview
Cubicle

A live pixel-art office for your AI agents — watch Paperclip, Claude Code, Codex and Gemini agents work, wait and fail in real time.

AI Specialized Tools
Muse Gadgets preview
Muse Gadgets

Meta's open-source kit for building physical AI gadgets that bring the Muse agent into the real world via ESP32 or Raspberry Pi.

AI Specialized Tools