Ltx 2
Official Python inference and LoRA trainer package for the Ltx 2 audio–video generative model.
Last verified:
What is Ltx 2?
Ltx 2 is a production-grade AI video generation model built for long-form, high-fidelity output with precise creative control. Developed by Lightricks, it generates synchronized audio and video in a single unified pass, supporting native 4K resolution at up to 50 FPS. The model is designed to support real production workflows rather than short experimental clips, making it suitable for professional studios, developers, and enterprise teams.
Key features include audio-to-video generation where voice, music, and sound effects define structure and pacing, 20-second long-form clip generation, depth-aware generation, OpenPose-driven motion control, precise camera control (dolly in/out/left/right), LoRA training support for custom styles and identities, detail upscaling, and video retake/extend capabilities. Ltx 2 offers two generation flows: Fast for rapid iteration and Pro for high-fidelity results. The model is available both as an API and as open-source weights on GitHub and Hugging Face.
Ltx 2 is ideal for professional video producers, VFX studios, developers building AI video applications, researchers, educators, and content creators who need predictable performance, deeper control over motion and camera behavior, and measurable specs for production environments.
Ltx 2 pricing
Pricing model: Freemium
LTX offers multiple pricing tiers: Free plan ($0/month) with 800 computing seconds one-time for initial testing; Lite plan at $15/month with 8,000 computing seconds/month for personal use; Standard plan at $35/month with 28,000 computing seconds/month including commercial use license; Pro plan at $125/month with 110,000 computing seconds/month, commercial use license, and 3 collaborators per project; Enterprise plans with custom pricing for unlimited access, privacy, and onboarding services. The LTX-2 API uses usage-based pricing billed per second of output video, with higher resolution and premium models (Pro vs Fast) having proportionally higher costs. Text-to-video, image-to-video, and audio-to-video endpoints all bill per second. HDR upscale is priced at $0.20/sec for up to 1080p and $0.40/sec for up to 1440p. Developers can experiment via the free LTX-2 Playground before committing to paid usage.
Ltx 2 pros
- Native 4K resolution at 50 FPS output
- Synchronized audio and video generation in one pass
- Up to 20-second long-form video generation
- Fast Flow for rapid iteration with lower compute
- Pro Flow for high-fidelity stable results
- Precise camera control (dolly in/out/left/right)
- Depth-aware generation for spatial understanding
- OpenPose-driven motion control for character animation
- LoRA training support for custom styles and identities
- Video retake feature for surgical editing without restart
- Scene extend capability for long-form continuity
- Detail upscaling tools for quality restoration
- Open-source weights available on GitHub and Hugging Face
- API access with stable, self-serve infrastructure
- Comparable ~18x faster step throughput than WAN 2.2 14B on H100
- Supports text-to-video, image-to-video, and audio-to-video
- ComfyUI and Fal integration for custom pipelines
- Distilled hybrid architecture for higher generation throughput
Ltx 2 cons
- Ultra mode 4K@50fps requires high-end GPUs or multi-GPU setups
- API pricing varies by resolution, frame rate, and usage volume
- Open-source version requires technical expertise for local deployment
- Maximum clip duration limited to 20 seconds per generation
- HDR upscale currently available only in 1080p resolution
- Full creative control requires detailed prompting and setup
- Local inference needs consumer-grade GPU minimum for Fast/Pro modes
- Enterprise plans require contacting sales for custom pricing
Frequently asked questions about Ltx 2
What is LTX-2?
LTX-2 is a professional-grade video generation model built for long-form, high-fidelity output with precise creative control. It is designed to support real production workflows rather than short experimental clips, generating synchronized audio and video at native 4K resolution up to 50 FPS.
What makes LTX-2 different from the LTXV model?
While the LTXV model focuses on creator-friendly generation and editing, LTX-2 is optimized for developers and production teams that require predictable performance, deeper control over motion and camera behavior, and measurable specs for production environments.
Does LTX-2 offer a downloadable model for local use?
Yes. LTX-2 is available for download in supported configurations for local and private deployment. The complete model weights, codebase, and tooling are publicly available for local and private use.
Is LTX-2 available as an open-source model?
Yes. The open-source version is available on GitHub and Hugging Face, enabling customization, fine-tuning, and local execution. This includes full model weights, a distilled version, several LoRAs, and a modular training framework.
How much does LTX-2's API cost?
Pricing for the LTX-2 API depends on usage, resolution, and frame rate. The API uses usage-based pricing billed per second of output video, with higher resolution and premium models (Pro vs Fast) having proportionally higher costs. Current plans are available on the LTX pricing page.
What video generation models are supported by LTX-2?
Supported variants include LTX-2-fast and LTX-2-pro, optimized for speed or maximum quality respectively. Fast Flow is built for rapid iteration with lower compute load, while Pro Flow delivers high-fidelity generation for stable, detailed results.
Does LTX-2 offer a free trial?
Yes. Developers can experiment with the model via the LTX-2 Playground, which allows testing prompts, exploring resolution and FPS settings, and previewing outputs before integrating the API into their stack.
Does LTX-2 integrate with ComfyUI or Fal?
Yes. LTX-2 supports integration with tools such as ComfyUI and Fal for custom pipelines and experimentation. Official ComfyUI integration is available through the ComfyUI-LTXVideo plugin.
What resolutions and FPS values does LTX-2 support?
LTX-2 supports resolutions of 1080p, 1440p, and 4K with frame rates of 24/25/48/50 FPS. Video duration can be up to 20 seconds depending on the model variant (ltx-2-fast or ltx-2-pro) used.
Does the LTX-2 API support image-to-video generation?
Yes. The LTX API includes image-to-video functionality, allowing developers to animate static images or evolve concepts into dynamic sequences using LTX-2. This is available through both sync and async endpoints.