Available now

LTX-2.3 22B: text to video

By Lightricks

LTX-2.3 22B is Lightricks' video model with every sampling control exposed: video and sound from a prompt or an image, with separate guidance for picture and audio.

Model ID
Maker
Lightricks

Input

Fields marked * are required. Review the price before generating.

Example prompt. Edit it or write your own.

Additional settings

Whether to generate audio for the video.

The Classifier-Free Guidance (CFG) scale for the video. Higher values result in more consistent and focused video content.

The Spatiotemporal Guidance (STG) scale for the video. Higher values result in more consistent and focused video content.

Controls the ratio between classifier-free guidance and spatiotemporal guidance.

Controls the ratio between video and audio modalities.

The Classifier-Free Guidance (CFG) scale for the audio. Higher values result in more consistent and focused audio content.

The Spatiotemporal Guidance (STG) scale for the audio. Higher values result in more consistent and focused audio content.

Controls the ratio between classifier-free guidance and spatiotemporal guidance.

Controls the ratio between video and audio modalities.

The gamma of gradient estimation during denoising. Set to 0 to disable.

The acceleration level to use.

The seed for the random number generator.

Whether to enable prompt expansion.

Set to 0 to disable.

The scale of the distill LoRA to use for the second and subsequent passes.

Price$0.01 / megapixel

You sign up first. Your settings are kept.

About this model

Overview

LTX-2.3 22B sets length by frame count, from 9 to 481 frames, and size by preset: square, square HD, or 4:3 and 16:9 in landscape or portrait. Sound is on by default and comes from the same prompt.

Every sampling setting is open: steps, scheduler, guidance for video and audio, multi-scale rendering and prompt expansion. You can save as MP4, WebM, ProRes 4444 MOV or GIF. The image mode adds a strength for how closely the clip keeps your picture.

What you can build

  • Clips with sound and a set frame count for an edit
  • ProRes MOV output for color grading
  • GIF loops for chat and docs
  • Repeatable tests with a fixed seed while you tune one setting

Prompting tips

  • Keep multi-scale on: it renders a small version first and uses it to guide the full-size clip.
  • Raise the audio guidance when the sound drifts from the prompt; raise the video guidance when the picture does.
  • The built-in negative prompt already blocks text, logos, cartoons and slow motion; edit it only if you want one of those.
  • In the image mode, lower the image strength to let the scene change more from your picture.

What to check

Check sound timing, repeated textures and object continuity across the entire clip.

Inputs

  • Prompt (required): The prompt to generate the video from.
  • Number of Frames (optional): The number of frames to generate. From 9 to 481.
  • Video Size (optional): The size of the generated video. Options: Square HD, Square, Portrait 4:3, Portrait 16:9, Landscape 4:3, Landscape 16:9.
  • Generate Audio (optional): Whether to generate audio for the video.
  • Use Multi-Scale (optional): Whether to use multi-scale generation. If True, the model will generate the video at a smaller scale first, then use the smaller video to guide the generation of a video at or above your requested size. This results in better coherence and details.
  • Number of Inference Steps (optional): The number of inference steps to use. From 8 to 50.
  • Video CFG Scale (optional): The Classifier-Free Guidance (CFG) scale for the video. Higher values result in more consistent and focused video content. From 1 to 20.
  • Video STG Scale (optional): The Spatiotemporal Guidance (STG) scale for the video. Higher values result in more consistent and focused video content. From 0 to 20.
  • Video Rescaling Scale (optional): The rescaling scale for the video. Controls the ratio between classifier-free guidance and spatiotemporal guidance. From 0 to 1.
  • Video Modality Scale (optional): The modality scale for the video. Controls the ratio between video and audio modalities. From 0 to 10.
  • Audio CFG Scale (optional): The Classifier-Free Guidance (CFG) scale for the audio. Higher values result in more consistent and focused audio content. From 1 to 20.
  • Audio STG Scale (optional): The Spatiotemporal Guidance (STG) scale for the audio. Higher values result in more consistent and focused audio content. From 0 to 20.
  • Audio Rescaling Scale (optional): The rescaling scale for the audio. Controls the ratio between classifier-free guidance and spatiotemporal guidance. From 0 to 1.
  • Audio Modality Scale (optional): The modality scale for the audio. Controls the ratio between video and audio modalities. From 0 to 10.
  • Gradient Estimation Gamma (optional): The gamma of gradient estimation during denoising. Set to 0 to disable. From 0 to 10.
  • Use Restart Sampling (optional): Whether to use restart sampling. This will inject a small amount of noise during each denoising step, which can help improve the quality of the generated video.
  • Scheduler (optional): The scheduler to use. Options: ltx2, Linear quadratic, beta.
  • Acceleration (optional): The acceleration level to use. Options: none, regular, high, full.
  • Negative Prompt (optional): The negative prompt to generate the video from.
  • Seed (optional): The seed for the random number generator.
  • Enable Prompt Expansion (optional): Whether to enable prompt expansion.
  • Video Output Type (optional): The output type of the generated video. Options: X264 (.mp4), VP9 (.webm), PRORES4444 (.mov), GIF (.gif).
  • Video Write Mode (optional): The write mode of the generated video. Options: fast, balanced, small.
  • Distill Lora First Pass Scale (optional): The scale of the distill LoRA to use for the first pass. Set to 0 to disable. From 0 to 1.
  • Distill Lora Second Pass Scale (optional): The scale of the distill LoRA to use for the second and subsequent passes. From 0 to 1.

Output and limits

  • Sound can be generated together with the video.
  • Readable text inside the video, especially small text, is unreliable.
  • Scenes with several interacting characters can lose consistency.
  • Longer clips can show more motion drift or scene changes.
  • Low-quality inputs cause glitches: heavy compression, collages, screenshots and text overlays.
  • Some prompts or outputs can be refused by safety checks.

Price guide

  • 1 megapixel: $0.01 / megapixel