TwelveLabs | Product Overview

Products

Enterprise

Research

Developers

Company

Playground

Talk to Sales

For everything you want to do with video.

Build features like semantic search and content recommenders, or novel applications that redefine what’s possible. Our state-of-the-art AI video understanding unlocks your video’s full potential.

Go to Playground

Human-level understanding.   For superhuman feats.

Experience semantic search and video-to-text capabilities that surpass anything you’ve tried before – video-native AI makes all the difference.

Find specific moments within your videos by describing the scene in natural language.

Generate

Generate text from videos - summary, chapters, highlights and more.

Embed

Build your own classifier using
natural language and run instantly.

Build your own classifier using natural language and run instantly.

Find any scene in natural language.

Fast, precise, context-aware results that truly understand what you’re looking for. Search across speech, text, audio and visuals to explore your video in every dimension.

Learn more

Try on Playground

Button Text

What do you want to find?

Try 'search' with a Sample App

See all sample apps

Python

Node

Olympic Video Classification Application

The Olympics Video Clips Classification Application is a powerful tool designed to categorize various Olympic sports using
video clips.

Try this sample app

Python

Node

Shade Finder App: Pinpoint Specific Colors in Videos

Whether you're looking to find the perfect berry-toned lipstick or just curious about spotting specific colors in your videos, this guide will help you leverage cutting-edge AI to do so effortlessly.

Try this sample app

Python

Node

Video Highlight Generator

The YouTube Chapter Highlight Generator is a tool developed to automatically generate chapter timestamps for YouTube videos.

Try this sample app

See all sample apps

Generate

Summarize, analyze and describe.

Create instant text formats like Q&As, hashtags, and summaries. Generate reports or get domain-specific analysis to answer your deepest questions.

Learn more

Try on Playground

Generate

Summarize, analyze and describe.

Create instant text formats like Q&As, hashtags, and summaries. Generate reports or get domain-specific analysis to answer your deepest questions.

Learn more

Try on Playground

Generate

Summarize, analyze and describe.

Create instant text formats like Q&As, hashtags, and summaries. Generate reports or get domain-specific analysis to answer your deepest questions.

Learn more

Try on Playground

Button Text

What do you want to describe?

Try 'generate' with a Sample App

See all sample apps

Node

Generate social media posts for your videos

This application simplifies the cross-platform video promotion workflow by generating unique posts for each social media platform.

Try this sample app

Node

Video Highlight Generator

This application automatically analyzes video content to create chapters and highlights, streamlining the video production workflow for content creators.

Try this sample app

Python

Node

Interview Analyzer

This application evaluates job interview performances using the ability of the Pegasus video understanding engine to generate text based on video content.

Try this sample app

See all sample apps

Embed

Turn video into vectors.

Build features like semantic search and domain classification with embeddings that capture nuance – and make the most
of all your data.

Learn more

Try on Playground

Embed

Turn video into vectors.

Build features like semantic search and domain classification with embeddings that capture nuance – and make the most
of all your data.

Learn more

Try on Playground

Embed

Turn video into vectors.

Build features like semantic search and domain classification with embeddings that capture nuance – and make the most
of all your data.

Learn more

Try on Playground

Button Text

What do you want to build?

Try 'embed' with a Sample App

See all sample apps

Node

Contextual and Personalized Ads

A tool for analyzing source footage, summarizing content, and recommending ads based on the footage's context and emotional tone.

Try this sample app

Python

Node

Recommendations using Multimodal Embeddings

Start exploring videos and discovering similar content powered by TwelveLabs Multimodal Embeddings.

Try this sample app

See all sample apps

Top Uses

Tailored for your industry.

Media & Entertainment

Advertising

Govt. & Security

Automotive

Automated Clip Generation

Create instant clips from longer content to use in social media and marketing.

Scene Selection

Quickly find and categorize key scenes. Choose the best takes and curate moments easily.

Bloopers and BTS Content

Get auto-curated highlight reels featuring the best of BTS or other special footage.

Automatic Tagging

Easily access and manage content in a vast video library — no manual tagging needed.

Content Summarization

Generate high-quality summaries and headlines to communicate the core message quickly.

Content Discovery

Find any key moment in footage easily — or help customers discover them too.

Real-Time Scene Classification

Get editorial support mid-workflow for seamless, swift production.

Ad Matching

Use contextual ad placement to ensure your customers’ attention and engagement.

Python

Node

from twelvelabs import TwelveLabs
import os

client = TwelveLabs("<YOUR_API_KEY>")

# Create new Index
index = client.index.create(        
     name="My First Index",
     engines=[
       {                
         "name": "marengo2.7",
         "options": ["visual", "audio"],
       },
     ],
  )
# Create new Task on Index (Upload the video)
video_path = os.path.join(os.path.dirname(__file__), "<YOUR_FILE_PATH>")
task = client.task.create(index_id=index.id, file=video_path, language="en")

# Wait for indexing to finishtask.wait_for_done()

# Search from your index
query = "An artist climbing up the ladder that he painted."
result = client.search.query(index.id, query, ["visual", "audio"])
‍print(result)

Integrate with your personalized SDK — and your vision.

Deploy your custom-trained model on any cloud. See and surface everything in your video, then go beyond with AI that can realize your most game-changing ideas.

Get the SDK

Top Uses

Tailored for your industry.

Media & Entertainment

Advertising

Govt. & Security

Automotive

Automated Clip Generation

Create instant clips from longer content to use in social media and marketing.

Scene Selection

Quickly find and categorize key scenes. Choose the best takes and curate moments easily.

Bloopers and BTS Content

Get auto-curated highlight reels featuring the best of BTS or other special footage.

Automatic Tagging

Easily access and manage content in a vast video library — no manual tagging needed.

Content Discovery

Find any key moment in footage easily — or help customers discover them too.

Content Summarization

Generate high-quality summaries and headlines to communicate the core message quickly.

Real-Time Scene Classification

Get editorial support mid-workflow for seamless, swift production.

Ad Matching

Use contextual ad placement to ensure your customers’ attention and engagement.

Python

Node

from twelvelabs import TwelveLabs
import os

client = TwelveLabs("<YOUR_API_KEY>")

# Create new Index
index = client.index.create(        
     name="My First Index",
     engines=[
       {                
         "name": "marengo2.7",
         "options": ["visual", "audio"],
       },
     ],
  )
# Create new Task on Index (Upload the video)
video_path = os.path.join(os.path.dirname(__file__), "<YOUR_FILE_PATH>")
task = client.task.create(index_id=index.id, file=video_path,
language="en")

# Wait for indexing to finish
task.wait_for_done()

# Search from your index
query = "An artist climbing up the ladder that he painted."
result = client.search.query(index.id, query, ["visual", "audio"])
‍print(result)

Integrate with your personalized SDK — and your vision.

Deploy your custom-trained model on any cloud. See and surface everything in your video, then go beyond with AI that can realize your most game-changing ideas.

Get the SDK

Top Uses

Tailored for your industry.

Media & Entertainment

Advertising

Govt. & Security

Automotive

Automated Clip Generation

Create instant clips from longer content to use in social media and marketing.

Scene Selection

Quickly find and categorize key scenes. Choose the best takes and curate moments easily.

Bloopers and BTS Content

Get auto-curated highlight reels featuring the best of BTS or other special footage.

Automatic Tagging

Easily access and manage content in a vast video library — no manual tagging needed.

Content Summarization

Generate high-quality summaries and headlines to communicate the core message quickly.

Content Discovery

Find any key moment in footage easily — or help customers discover them too.

Real-Time Scene Classification

Get editorial support mid-workflow for seamless, swift production.

Ad Matching

Use contextual ad placement to ensure your customers’ attention and engagement.

Python

Node

from twelvelabs import TwelveLabs
import os

client = TwelveLabs("<YOUR_API_KEY>")

# Create new Index
index = client.index.create(        
     name="My First Index",
     engines=[
       {                
         "name": "marengo2.7",
         "options": ["visual", "audio"],
       },
     ],
  )
# Create new Task on Index (Upload the video)
video_path = os.path.join(os.path.dirname(__file__), "<YOUR_FILE_PATH>")
task = client.task.create(index_id=index.id, file=video_path,
language="en")

# Wait for indexing to finish
task.wait_for_done()

# Search from your index
query = "An artist climbing up the ladder that he painted."
result = client.search.query(index.id, query, ["visual", "audio"])
‍print(result)

Integrate with your personalized SDK — and your vision.

Deploy your custom-trained model on any cloud. See and surface everything in your video, then go beyond with AI that can realize your most game-changing ideas.

Get the SDK

Play for free.
Pay as you go.

Our tiered pricing lets you play and build, then launch and grow. Start with one of our foundational models and pay only for what you use.

Free

For testing and building

Get started

Indexing limit

<10 hours

Environment

Shared

Cloud integration

Org account

SSO/ SAML

Finetune

Developer

For launching and growing

Upgrade

Indexing limit

<10k hours

Environment

Shared

Cloud integration

Included

Org account

SSO/ SAML

Finetune

Enterprise

For scaling and services

Learn more

Indexing limit

Unlimited

Environment

Dedicated

Cloud integration

Included

Org account

Included

SSO/ SAML

Included

Finetune

Included

Our stable of models.

These state-of-the-art video foundation models are setting the standard for video intelligence.

Models

At TwelveLabs, we’re developing video-native AI systems that can solve problems with human-level reasoning. Helping machines learn about the world — and enabling humans to retrieve, capture, and tell their visual stories better.

Marengo 2.7

Sets new benchmarks in zero-shot text-to-video, text-to-image, and text-to-audio retrieval tasks with a single embedding model.

Outperforms Google's VideoPrism-G model by +10% on the MSR-VTT dataset and +3% on the ActivityNet dataset

Surpasses the SOTA image foundation model in zero-shot text-to-image retrieval tasks, showcasing its ability to understand and process visual content.

Marengo 2.7

Sets new benchmarks in zero-shot text-to-video, text-to-image, and text-to-audio retrieval tasks with a single embedding model.

Outperforms Google's VideoPrism-G model by +10% on the MSR-VTT dataset and +3% on the ActivityNet dataset

Surpasses the SOTA image foundation model in zero-shot text-to-image retrieval tasks, showcasing its ability to understand and process visual content.

Marengo 2.7

Sets new benchmarks in zero-shot text-to-video, text-to-image, and text-to-audio retrieval tasks with a single embedding model.

Outperforms Google's VideoPrism-G model by +10% on the MSR-VTT dataset and +3% on the ActivityNet dataset

Surpasses the SOTA image foundation model in zero-shot text-to-image retrieval tasks, showcasing its ability to understand and process visual content.

Pegasus 1.2

Processes the video input to generate rich embeddings from both video frames and audio speech recognition (ASR) data.

Maps the video embeddings to corresponding language embeddings, creating a shared space where video and text representations are aligned.

The large language model decoder takes the aligned embeddings and user prompts to generate coherent and contextually relevant text output.

Pegasus 1.2

Processes the video input to generate rich embeddings from both video frames and audio speech recognition (ASR) data.

Maps the video embeddings to corresponding language embeddings, creating a shared space where video and text representations are aligned.

The large language model decoder takes the aligned embeddings and user prompts to generate coherent and contextually relevant text output.

Pegasus 1.2

Processes the video input to generate rich embeddings from both video frames and audio speech recognition (ASR) data.

Maps the video embeddings to corresponding language embeddings, creating a shared space where video and text representations are aligned.

The large language model decoder takes the aligned embeddings and user prompts to generate coherent and contextually relevant text output.

Models

We’re fundamentally transforming how people see, experience and use video through AI.

Grounded in research and knowledge sharing, we’re learning and growing alongside our partners, with consideration for the future.

Our Research

A Joint Sequence Fusion Model for Video Question Answering and Retrieval

ECCV 2018

1st place in LSMDC challenge in ICCV 2017

AdaCoF: Adaptive Collaboration of Flows for Video Frame Interpolation

CVPR 2020

Character Grounding and Re-Identification in Story of Videos and Text Descriptions

ECCV 2020

1st place in LSMDC challenge in ICCV 2019

Psynet: Self-supervised Approach to Object Localization using Point Symmetric Transformation

AAAI 2020

Tutorials

Take the reins with our quick-start tutorials.

Twelve Labs Secures $30 Million in Funding, Validating the Importance of Twelve Labs' Video Understanding Technology to the AI Ecosystem

Twelve Labs is building AI that can analyze and search through videos

The State of Video-Language Models: Research Insights from the Inaugural NeurIPS Workshop

Our Series A to Build the Future of Multimodal AI

Twelve Labs Secures $30 Million in Funding, Validating the Importance of Twelve Labs' Video Understanding Technology to the AI Ecosystem

Twelve Labs is building AI that can analyze and search through videos

The State of Video-Language Models: Research Insights from the Inaugural NeurIPS Workshop

Our Series A to Build the Future of Multimodal AI

See all tutorials

Ready to see your video differently?

Try your own video in our Playground to see next-level video intelligence in action.

Try on Playground

Talk to Sales

Try on Playground

Talk to Sales

Ready to see your video differently?

Try your own video in our Playground to see next-level video intelligence in action.

Try on Playground

Talk to Sales

Try on Playground

Talk to Sales

Product

Social

For Developers

Solutions

Government & Security

Media & Entertainment

Company

Legal

2025

Product

Social

Solutions

Government & Security

Media & Entertainment

For Developers

Legal

Company

2025

Product

Social

For Developers

Solutions

Government & Security

Media & Entertainment

Company

Legal

2025

For everything you want to do with video.

Human-level understanding. For superhuman feats.

Find any scene in natural language.

Find any scene in natural language.

Find any scene in natural language.

What do you want to find?

Olympic Video Classification Application

Shade Finder App: Pinpoint Specific Colors in Videos

Video Highlight Generator

Summarize, analyze and describe.

Summarize, analyze and describe.

Summarize, analyze and describe.

What do you want to describe?

Generate social media posts for your videos

Video Highlight Generator

Interview Analyzer

Turn video into vectors.

Turn video into vectors.

Turn video into vectors.

What do you want to build?

Contextual and Personalized Ads

Recommendations using Multimodal Embeddings

Tailored for your industry.

Integrate with your personalized SDK — and your vision.

Tailored for your industry.

Integrate with your personalized SDK — and your vision.

Tailored for your industry.

Integrate with your personalized SDK — and your vision.

Play for free.Pay as you go.

Free

Free

Developer

Developer

Enterprise

Enterprise

Our stable of models.

Marengo 2.7

Marengo 2.7

Marengo 2.7

Pegasus 1.2

Pegasus 1.2

Pegasus 1.2

We’re fundamentally transforming how people see, experience and use video through AI.

Take the reins with our quick-start tutorials.

Ready to see your video differently?

Ready to see your video differently?

Human-level understanding.   For superhuman feats.

Play for free.
Pay as you go.