Lightricks LTX-2.5 Fast

Fast text-to-video and image-to-video with synchronized audio. Portrait and landscape, up to 10s at 720p/1080p.

Cost: 78 credits

Input

Try:

Write one continuous present-tense paragraph. Include camera movement, lighting, and atmosphere.

Upload an image to animate (leave empty for text-to-video)

Optional. When set, the image is the first frame (image-to-video). Leave empty for text-to-video.

720p is cheaper; 1080p has more detail. Longer clips cost more at higher resolution.

Clip length in seconds. Credits scale with duration and resolution.

16:9 landscape or 9:16 portrait.

Generate synchronized audio with the video.

Output

Generated content will appear here

Example Results

Video Generation:

A calm ocean wave rolls toward a sandy beach at sunset, soft golden light, gentle camera pan left, ambient waves and wind.

Frequently Asked Questions

What is Lightricks LTX-2.5 Fast?
LTX-2.5 Fast is Lightricks’ speed-optimized video model on Replicate. It generates MP4 clips with optional synchronized audio from a text prompt, and can also animate a first-frame image into video.
Can I do both text-to-video and image-to-video?
Yes. Leave the first-frame image empty for text-to-video. Upload an image to use it as the starting frame for image-to-video.
What resolutions and durations are available?
This page offers 720p and 1080p, with durations of 6, 8, or 10 seconds. Aspect ratio can be 16:9 (landscape) or 9:16 (portrait).
Does the video include sound?
When Generate Audio is enabled (default), the model produces audio aligned with the motion and pacing of the clip.
How should I write prompts?
Use one continuous paragraph in present tense. Be specific about camera angles, lighting, movement, and atmosphere. Longer, detailed prompts usually work better than short keyword lists.
What format do I get back?
An MP4 video URL suitable for download and sharing. Audio is included when Generate Audio is on.
How is this different from LTX-2 Fast on this site?
LTX-2.5 Fast is the newer Replicate model with a single endpoint for both text-to-video and image-to-video, plus resolution and aspect-ratio controls on this page.