Natural-language video search
Ask for a person, place, scene, quote, or moment in plain language and Jockey returns matching clips from a connected video library.
Overview
Jockey is TwelveLabs’ video intelligence agent. The homepage describes it as an agent layer for video understanding that can organize, search, and assemble across a video library, with the product currently in research preview.
The product is positioned around natural-language interaction with video: users can ask questions, find specific clips or quotes, and generate summaries or highlight reels. The page also shows Jockey connected to Claude, with a developer path through the TwelveLabs API and SDK, and links the product to TwelveLabs’ Marengo and Pegasus models for retrieval and video-language reasoning.
Core capabilities shown on the site
Clip-level retrieval
Use the assistant to move from broad questions to specific moments without browsing folders or opening every file.
Instant highlight reel assembly
Point Jockey at a topic and receive an edit-ready cut of relevant moments, as shown in the highlight reel examples.
Assistant-based workflow
Connect Jockey to Claude and ask questions directly from the chat interface you already use; the page notes ChatGPT is coming soon.
API and SDK access
Use the developer examples to query the product through the TwelveLabs SDK with a model name, knowledge store ID, and session ID.
Timestamped outputs
Return machine-readable, timestamped metadata and frame-accurate results for downstream products and playback links, based on the broader platform pages.
Practical use cases shown on the page
Marketing creative review
Search a creative archive for ads by hook, talent, logo, format, or tone, then compare patterns across the library before planning the next shoot.
Creator highlight extraction
Ask for the strongest moments in a footage archive and turn them into a rough, edit-ready highlight reel.
Archive retrieval
Find a quote, a specific scene, or a named person in a large media archive without opening files one by one.
Developer integration
Embed video understanding into a product search or analytics workflow using the API examples and knowledge-store pattern shown on the page.
Assistant-native library search
Ask for a person, place, scene, or moment directly inside Claude after connecting the library through the integration workflow shown on the homepage.
Pros and Cons
Pros
- Searches video by meaning and natural-language questions instead of filenames or manual browsing.
- Supports multiple workflows shown on the site, including quote finding, clip retrieval, summaries, and highlight reel creation.
- Can be used through Claude and through developer-facing SDK/API patterns.
- Built on TwelveLabs models that are described separately for retrieval, embeddings, and video-language analysis.
Cons
- Jockey is in research preview and the homepage says it is accepting limited sign-ups.
- The source shows ChatGPT support as coming soon rather than available today.
- The public page does not provide a full, detailed specification of supported workflows, permissions, or integrations beyond the examples shown.
FAQ
Is Jockey generally available?
Jockey is presented as a research preview. The homepage says it is currently accepting limited sign-ups and offers a free-trial style entry point, while the pricing page shows TwelveLabs also has Free, Developer, and Enterprise API plans for its broader platform.
What does Jockey do?
The homepage says Jockey can organize, search, and assemble across a video library. The examples show natural-language search, quote finding, highlight reel creation, and analysis that returns clip-level results.
Which assistants or integrations are shown?
The homepage shows two named assistant integrations: Claude is connected now, and ChatGPT is listed as coming soon. It also shows an MCP endpoint for Jockey.
Is Jockey only a chat interface or can developers use it in code?
The page shows Jockey being used through a chat-style interface where a user can ask for clips, summaries, and lists of moments. It also includes SDK example code for calling a Jockey model via the TwelveLabs client.
What models does Jockey use?
The source specifically ties Jockey to TwelveLabs models Marengo and Pegasus. The model pages describe Marengo as the retrieval and embedding model and Pegasus as the video-language model for description and analysis.
Quick Facts
- Category
- Video intelligence agent
- Primary use
- Search, organize, and assemble content from video libraries
- Interface
- Chat-style assistant plus API/SDK
- Shown integrations
- Claude; ChatGPT coming soon
- Underlying models
- Marengo and Pegasus
- Website
- twelvelabs.io
Jockey 대안
RLAMA
RLAMA is a local AI platform for building RAG systems and intelligent agents on macOS, Linux, and Windows, with HTTP API support.
Crossing Minds
Crossing Minds is a retrieval platform for accurate, secure, scalable AI applications. The public site also says the team is joining OpenAI.
Orca
Orca is an Agent Development Environment for shipping with coding agents, running multiple CLI agents in parallel across isolated worktrees, with desktop and mobile workflows.
AI Magicx
AI Magicx is a unified AI workspace for chat, image, video, voice, music, email and developer tasks, helping teams and creators manage multiple models in one place.
Paper
Paper is a design tool that connects canvas, code, and AI agents so teams can create, share, and ship work in one workflow. Includes desktop app and MCP access.
blop
blop is a QA agent that writes browser tests as code in your repo, runs them in CI, clusters repeated failures, and can open PRs to fix broken tests.