LTX 2 - Production-Grade AI Video GenerationLTX 2 by Lightricks: controllable, high-fidelity video generation with open weights

LTX 2 is a DiT-based audio-video foundation model designed to generate synchronized video and audio in a single model. Built for real production workflows, LTX 2 emphasizes precise creative control over structure, motion, and camera behavior—so outputs look intentional, not random. With open weights for local execution and API-ready deployment, LTX 2 supports modern pipelines across text-to-video and image-to-video, plus upscalers for higher resolution and frame rate when you need it.

Why Choose LTX 2
LTX 2 makes controllable video generation production-ready

LTX 2 combines high-fidelity generation with measurable specs and real creative control. Direct structure, motion, and camera intent with LTX 2 controls, then scale quality with LTX 2 upscalers and production configurations—whether you run locally with open weights or integrate via API.

Feature

LTX 2 Foundation Model

LTX 2 is a diffusion-based audio-video foundation model from Lightricks. LTX 2 is built to generate video and audio together with coherent timing, and it supports both text-to-video and image-to-video workflows.

Audio + Video, Synchronized

LTX 2 is designed to keep motion, dialogue, ambience, and music aligned to the visuals. When you need synchronized sound and picture, LTX 2 is built to generate them in one coherent process.

Precise Control with LTX 2

LTX 2 is optimized for predictable results and deeper control. Guide structure, depth, pose, edges, and camera behavior so LTX 2 outputs follow your intent across complex scenes.

Production Specs and Upscalers

LTX 2 supports production-oriented configurations, including variants optimized for speed or maximum quality. LTX 2 upscalers extend resolution and frame rate for sharper visuals and smoother motion when your workflow demands it.

LTX 2 FAQ

Everything you need to know about LTX 2

Answers to common questions about LTX 2: what it is, how to control it, and how to use LTX 2 in production workflows.

LTX 2 is an audio-video foundation model from Lightricks. LTX 2 is designed for production-grade generation with open weights and a focus on controllable, high-fidelity video creation.
LTX 2 supports modern multimodal workflows, including text-to-video and image-to-video. LTX 2 is also designed to generate synchronized audio and video within a single model.
LTX 2 is built around controllability. Use LTX 2 control options (like depth, pose, and edge guidance) to steer structure and motion, then add detail enhancement and upscaling to push final quality.
Yes. LTX 2 has open weights and is available for local execution in supported configurations, with documentation and integration paths for common tooling and pipelines.
LTX 2 includes multiple checkpoints and variants, including full and distilled options, plus spatial and temporal upscalers. LTX 2 variants let you trade off speed, quality, and compute based on your production needs.
For best results with LTX 2, use video sizes divisible by 32 and frame counts that follow the model’s recommended constraints. Clear prompts and strong control signals help LTX 2 deliver stable structure and motion.
LTX 2 can be used through developer APIs and popular community tools. LTX 2 also supports open-source integrations for building custom pipelines, automation, and repeatable production workflows.

Still have questions? Contact our support team

Limited Time Offer

Start Creating with LTX 2 Today

Bring LTX 2 control to your next video

Use LTX 2 to generate high-quality video with production-ready control. Start with a prompt or a reference image, steer motion and structure, then refine with LTX 2 enhancements for results that hold up in real creative pipelines.

  • Generate with LTX 2 from text or an image reference
  • Guide structure, motion, and camera intent with LTX 2 controls
  • Scale quality with LTX 2 upscalers and detail enhancement
  • Use LTX 2 in local or API-driven production workflows