LTX-2 Pro

LTX-2 Pro Product Information

What is LTX-2 Pro?

LTX-2 Pro is a 19-billion-parameter open-source video foundation model that creates synchronized, high-resolution video and audio in a single generation pass. Comprising 14 billion parameters for video and 5 billion for audio, it generates native 4K (3840×2160) video at up to 50 FPS, with clips lasting up to 20 seconds. It supports a range of multimodal inputs, including text prompts, images, depth maps, keyframes, and short reference videos. The model features integrated audio synthesis—producing dialogue, ambient sounds, music, and Foley effects perfectly aligned with the visual content. Optimized for efficient inference with NVFP4/NVFP8, it is released under the Apache 2.0 license, allowing teams to download the model weights for local fine-tuning and deployment, or use the hosted web generator (which requires credits).

Who will use LTX-2 Pro?

• Filmmakers and VFX artists
• Advertising agencies and marketing teams
• Social media content creators (for TikTok, Reels, Shorts)
• Pre-visualization and concept artists
• Independent studios and rapid prototyping teams
• AI researchers and model engineers

How to use the LTX-2 Pro?

• Step 1: Go to https://z-image.io and sign in or create an account (an account is required to generate videos).
• Step 2: Open the LTX-2 generator, select the model mode (Fast or Pro) and your preferred generation mode (Text-to-Video or Image-to-Video).
• Step 3: Choose your resolution and duration (note: clips longer than 10 seconds must use 1080p; native 4K is supported for clips up to 20 seconds). Enable audio if desired, and optionally upload images or keyframes for additional conditioning.
• Step 4: Enter a detailed prompt, use prompt enhancement tools if needed, then submit your generation job (web generation consumes credits).
• Step 5: Check your results in the Recent Tasks section, refine your prompts or settings, and download the final video for use or further editing.

Platform

• Web

LTX-2 Pro's Core Features & Benefits

The Core Features

• Synchronized single-pass audio and video generation
• Native 4K video output at up to 50 FPS
• Multimodal conditioning: text, images, video, depth maps, keyframes
• 19-billion-parameter DiT architecture (14B video + 5B audio)
• Open-source under Apache 2.0 license with full model weights and code
• Text-to-Video and Image-to-Video generation modes
• Optimized for efficient inference (NVFP4/NVFP8)

The Benefits

• Cinematic-quality videos with perfectly synchronized audio
• Commercial use permitted via the Apache 2.0 license
• Flexible deployment: use the hosted web generator or run locally with model weights
• Accelerates prototyping and lowers production costs
• Precise creative control through multimodal inputs and keyframe conditioning

LTX-2 Pro's Main Use Cases & Applications

• Film pre-visualization and VFX concept generation
• Advertising and promotional video creation
• Short-form content for social media (TikTok, Reels, Shorts)
• Campaign A/B testing and rapid concept iteration
• Mood films and cinematic trailers
• Prototype and UI/UX motion previews

LTX-2 Pro's Pros & Cons

The Pros

Open-source Apache 2.0 license allows for commercial use

Produces high-quality native 4K video at up to 50 FPS with synchronized audio

Accepts multimodal inputs for precise and controllable generation

Full model and pipeline available for local customization and fine-tuning

Optimized for performance on consumer-grade GPUs

The Cons

Maximum clip length is limited (6–20 seconds), not suitable for long-form video

High computational power is needed for local 4K inference

The web generator requires credits for video creation

Generated output may need post-processing for professional production use

No direct mobile applications or Chrome extension is currently listed

LTX-2 Pro's Pricing

Details of Pricing Plan

Z-Image Basic

7.99 USD

• 360 Credits
• ~180 images/month
• $0.044 per image
• More relaxed generation limits
• Standard generation speed
• Email support

Z-Image Pro

25.9 USD

• 1320 Credits
• ~660 images/month
• $0.039 per image
• More relaxed generation limits
• Priority processing
• Private task generation (self-delete account)
• Priority support
• Commercial image license

Z-Image Max

59.9 USD

• 3600 Credits
• ~1800 images/month
• $0.033 per image
• More relaxed generation limits
• Fastest generation speed
• Permanent image storage
• Full commercial rights
• Private task generation (self-delete account)

Z-Image Ultra

99.9 USD

• 7500 Credits
• ~3750 images/month
• $0.026 per image
• More relaxed generation limits
• Fastest generation speed
• Permanent image storage
• Full commercial rights
• Private task generation (self-delete account)

Discount:

New Year Sale 50% OFF

For the latest prices, please visit: https://z-image.io/pricing

FAQs of LTX-2 Pro

What is LTX-2?

LTX-2 is an open-source AI video foundation model developed by Lightricks. With 19 billion parameters, it generates synchronized 4K video and audio in one pass using a Diffusion Transformer (DiT) architecture.

Is LTX-2 open source and what license does it use?

Yes, it is fully open-source. LTX-2 is released under the permissive Apache 2.0 license, granting access to model weights, training code, and the inference pipeline for both personal and commercial use.

What resolutions and frame rates are supported?

LTX-2 supports native 4K output (3840×2160) at up to 50 frames per second. Lower resolutions, such as 1080p, are also available and are required for generating longer videos on the hosted platform.

How long can generated videos be?

Generated video clips can be between 6 and 20 seconds long. When using the hosted web service, clips longer than 10 seconds must be generated at 1080p resolution.

Does LTX-2 produce audio as well as video?

Yes. LTX-2 generates synchronized audio—including dialogue, ambient sounds, music, and Foley effects—in the same pass as the video, ensuring perfect alignment.

Can I use videos created with LTX-2 commercially?

Yes. The Apache 2.0 license permits commercial use of the generated videos. Please review any specific terms of service if you are using the hosted web generation service which requires credits.

How do I generate a video on the hosted site?

Sign in to your account at z-image.io, navigate to the LTX-2 generator, select your model, mode, resolution, and duration. Enter a descriptive prompt (or upload images), enable audio if desired, and submit the job. Note that generation consumes credits.

Are model weights and code available for local use?

Yes. As an open-source project under Apache 2.0, the complete model weights and code are available for download, allowing for local deployment, fine-tuning, and integration into custom pipelines.

What inputs can I use to control generation?

You can guide generation using detailed text prompts, images, short reference video clips, depth maps, and multi-keyframe conditioning for precise temporal and stylistic control.

Where can I find pricing and credits information?

Detailed pricing and information about credit packs for the hosted web generator are available at https://z-image.io/pricing.

LTX-2 Pro Company Information

• Z-Image AI
• https://z-image.io

• [email protected]

LTX-2 Pro Screenshot