ltx-2

Generate synchronized audio-video content from text prompts or source media using LTX-2 diffusion pipelines.

1|Updated Apr 3, 2026
One-click install
npx skills add https://github.com/bobvarkey/openclaw-workspace --skill ltx-2
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ltx-2
Source: https://github.com/bobvarkey/openclaw-workspace/tree/main/skills/local/ltx-2
Command: npx skills add https://github.com/bobvarkey/openclaw-workspace --skill ltx-2

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill produces high-quality generated video when a user needs text-to-video, image-to-video, video-to-video, audio-driven video, or keyframe interpolation without building a custom diffusion pipeline.

Core Features & Use Cases

  • Text-to-Video Generation: Create synchronized audio and video from a prompt using the recommended two-stage pipeline.
  • Image and Video Transformation: Animate still images, transform existing videos with IC-LoRA, and regenerate specific time regions.
  • Production-Oriented Pipelines: Choose between distilled, one-stage, two-stage, audio-to-video, and keyframe interpolation workflows depending on quality and speed needs.
  • Use Case: A creator can turn a product concept image into a polished motion clip with sound, or rapidly prototype multiple prompt variations for review.

Quick Start

Activate the LTX-2 environment, ensure the required models are in place under the models directory, and ask for the desired video generation workflow with your prompt, input media, and output path.

Frequently Asked Questions about ltx-2

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

FAQPage Schema
How do I generate synchronized audio-video content from text prompts?

Image-to-video generation with LTX-2 animates still images by applying the diffusion pipeline to your source media. You can transform existing videos using IC-LoRA and regenerate specific time regions by specifying the desired input media and output path.

Can I use audio to drive video generation?

LTX-2 offers distilled, one-stage, two-stage, audio-to-video, and keyframe interpolation workflows. Choose the distilled or one-stage pipeline for rapid prototyping, and the two-stage pipeline for higher quality synchronized audio-video output.

What do I need to run text-to-video diffusion pipelines?

Running text-to-video diffusion pipelines requires Python 3, GPU-capable runtime support, and the appropriate model checkpoints, Gemma text encoder, and upscaler or LoRA assets placed under the models directory. Ensure these prerequisites are met before activating the environment.

Does the video generation workflow support keyframe interpolation?

Yes, keyframe interpolation is a dedicated workflow supported by LTX-2. It allows you to generate intermediate frames between set keyframes, providing a production-oriented approach for smooth video animation and specific time region regeneration.