What is LibTV Agent
LibTV Agent is a product under LiblibAI, LibTV launched as a professional-gradevideo creationagent, positioned as the world's largest professional video Skill Hub. The product breaks through the traditional workflow stitching model and builds a unified baseline around finished-film capabilities. It can autonomously complete creative understanding, shot planning, video generation, and finished-film optimization, achieving a first-generation success rate of over 80% for video tasks within 3 minutes, enabling finished-product-grade direct output for scenarios such as commercial advertising, short dramas, and film and television.
Main features of LibTV Agent
-
High-baseline all-around video Agent: Builds a unified baseline around finished-film capabilities, autonomously completing creative understanding, shot planning, generation, and optimization, rather than single-shot stitching.
-
Director-level Video Skill Hub: Launches more than one hundred Skills in the first batch, covering film and television directing, brand advertising, e-commerce marketing, short drama production, and self-media creation, encapsulating complete creative capabilities.
-
Dual-view creation: Simultaneously outputs Storyboard and Node Workflow, supporting free switching between natural language creation and node-level workflows.
-
Agent-driven editing: Editing is embedded as an Agent capability, supporting continuous completion of editing, restructuring, packaging, multilingual localization, and detail optimization through natural language.
How to use LibTV Agent
- Visit the official website: Visit LibTV official website and enter the LibTV Agent interface.
-
Select a Skill: After entering the platform, choose a professional Video Skill just like choosing a director, and determine the creative style and field.
-
Input ideas: Describe creative requirements in natural language, and the Agent automatically understands and plans shots and narrative structure.
-
Generate finished film: The Agent autonomously invokes model combinations, prompting strategies, and style control to complete video generation and preliminary optimization.
-
Collaborative refinement: Continuously modify visuals, pacing, subtitles, voiceover, and packaging through natural language, and the Agent reconstructs the edit in real time.
-
Multi-platform publishing: Export final versions in different languages and platform specifications with one click, completing global distribution.
Core advantages of LibTV Agent
-
Finished-product-grade direct output: For videos within 3 minutes, the one-generation success rate for reaching deliverable standards exceeds 80%, saying goodbye to clip stitching.
-
Encapsulation of professional directing experience: In collaboration with 200+ professional creators, distilling Wes Anderson's cinematic aesthetics, King Hu's classical wuxia aesthetics, and more into callable Agent Skills.
-
Reusable creation process: Not only generates content, but also generates a complete creation process, supporting multi-Session collaboration and long-range memory for a single project.
-
One-click multi-platform delivery: Supports one-click generation of versions for different languages, regions, and platforms, achieving create once, publish globally.
Comparison of LibTV Agent with similar competing products
| Dimension | LibTV Agent | Vidu Agent(Shengshu Technology) |
|---|---|---|
| Product positioning | The world's largest professional video Skill Hub, oriented toward finished-product-grade direct output across all scenarios such as film and television, advertising, short dramas, and self-media | Commercially usable one-stop AI video Agent, focused on one-click finished-film generation in the advertising and e-commerce fields |
| Output Target | Over 80% one-time generation success rate for videos within 3 minutes, delivering finished, production-grade complete videos | Complete 15–30 second ad films/MVs, ready for direct commercial use |
| Creation Mode | A unified Agent autonomously completes the entire workflow of creative understanding, shot planning, generation, editing, and packaging | Multi-agent collaboration (director, storyboard, generation, editing), automatically breaking down selling points and scene templates |
| Skill / Experience Encapsulation | 100+ director-level Video Skills; 200+ professional creators distill aesthetic styles and narrative pacing into reusable Agent capabilities | Built-in advertising law compliance library, platform rule engine, and 50+ scene templates, with a focus on e-commerce advertising experience |
| Video Duration | Supports long video creation within 3 minutes, covering short dramas and film/TV previsualization | Currently only supports two specifications, 15 seconds and 30 seconds; long video features are in development |
| Views and Collaboration | Dual-view output of Storyboard and Node Workflow, supporting multi-Session collaboration and long-range memory within a single project | Primarily one-click video creation, multi-image reference (up to 7 images) to maintain subject consistency, enterprise edition supports team spaces |
| Editing Capabilities | Agent-driven editing, continuously modifying visuals, pacing, subtitles, and voiceover via natural language, generating multilingual/multi-platform versions | Editing and compositing agents automatically complete shot stitching, transitions, and audio-visual synchronization, supporting brand VI constraints |
Application scenarios of LibTV Agent
-
Commercial Advertising: Brand owners can quickly invoke advertising creative Skills to generate finished ad films that match director-level pacing and brand tonality.
-
Short Drama Production: Short drama teams use professional narrative Skills to complete storyboard design, shot generation, and editing packaging, shortening the production cycle.
-
Film/TV Previsualization: Directors use the Storyboard view to quickly output visualized previz, assisting investor decisions and team communication.
-
Self-Media Creation: Creators invoke self-media-specific Skills to generate multi-platform adapted versions with one click, improving content output efficiency.
-
E-commerce Marketing: Merchants use e-commerce marketing Skills to batch-generate product showcase videos and automatically complete multilingual localized distribution.