ltx2

Create short video clips from prompts or images using LTX-2.3.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/hanzoai/video --skill ltx2-hanzoai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ltx2
Source: https://github.com/hanzoai/video/tree/main/.agents/skills/ltx2
Command: npx skills add https://github.com/hanzoai/video --skill ltx2-hanzoai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Produces short, cinematic motion clips from text prompts or input images using the LTX-2.3 DiT model, accelerating video production and concept testing without manual editing.

Core Features & Use Cases

  • Text-to-video: create cinematic clips from descriptive prompts, defaulting to ~5 seconds.
  • Image-to-video / motion: animate still images or generate motion-rich b-roll content.
  • Custom parameters: control resolution, frame count, and output format to fit project specs.
  • Use Case: draft engaging social posts, product demos, or brand intros with consistent visual style.

Quick Start

Generate a 5-second cinematic clip from a descriptive prompt using the LTX-2.3 model.

Frequently Asked Questions about ltx2

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate cinematic video clips from text prompts?

To generate cinematic video clips from text prompts, provide a descriptive prompt to the text-to-video pipeline. The system creates short, reproducible clips defaulting to ~5 seconds using the LTX-2.3 model.

Can I animate still images to create motion-rich b-roll?

Yes, you can animate still images to create motion-rich b-roll using the image-to-video workflow. This feature animates input images to produce cinematic motion clips without manual editing.

How do I control resolution and duration for text-to-video generation?

You control resolution and duration by configuring custom parameters within the prompt-driven pipeline. This allows you to adjust frame count and output format to fit specific project specs.

Do I need a specific endpoint deployed to use LTX-2 for video production?

Yes, you need a deployed LTX-2.3 endpoint on Modal A100-80GB. This environment is required to run the prompt-driven pipeline and generate reproducible video clips.

What are the limitations when generating short video clips with LTX-2?

The system produces short clips defaulting to ~5 seconds and requires a deployed Modal A100-80GB endpoint. It accelerates concept testing but focuses on short-form output rather than long video sequences.