replicate/cog-ltx-video-0.9.7-distilled

replicate/cog-ltx-video-0.9.7-distilled

Description: Cog wrapper for ltx-video-0.9.7-distilled

Language: Python

Stars: 2

Forks: 1

Open issues: 0

Created: 2025-07-11T20:04:36Z

Pushed: 2025-07-24T20:32:45Z

Default branch: main

Fork: no

Archived: no

README:

LTX Video 0.9.7 Distilled - Replicate Cog

A high-quality video generation model that creates videos from text prompts, images, or existing videos using the LTX Video 0.9.7-distilled architecture.

Features

• Text-to-Video: Generate videos from text descriptions
• Image-to-Video: Animate static images into videos
• Video-to-Video: Transform existing videos with new prompts
• Multiple aspect ratios: 16:9, 1:1, 9:16, or match input image
• Flexible resolution: 480p or 720p output
• Fast mode: 20-40% faster generation with optimized settings

Usage

Text-to-Video

cog predict -i prompt="A cat playing with a ball of yarn"

Image-to-Video

cog predict -i prompt="A cute little penguin takes out a book and starts reading it" -i image=@peng.png

Video-to-Video

cog predict -i prompt="The video depicts a winding mountain road covered in snow, with a single vehicle traveling along it. The road is flanked by steep, rocky cliffs and sparse vegetation. The landscape is characterized by rugged terrain and a river visible in the distance. The scene captures the solitude and beauty of a winter drive through a mountainous region" -i video=@cosmos.mp4

Key Parameters

• prompt: Text description of the desired video
• resolution: Output height in pixels (480 or 720)
• aspect_ratio: Video dimensions (16:9, 1:1, 9:16, or match_input_image)
• num_frames: Number of frames to generate (9-257, default: 121)
• fps: Frames per second for output video (default: 24)
• go_fast: Enable fast mode for quicker generation (default: true)

Output

Returns an MP4 video file with the generated content at the specified resolution and frame rate.

Model Info

• Based on LTX Video 0.9.7-distilled by Lightricks
• Includes spatial upsampler for enhanced quality
• Optimized for memory efficiency and fast inference
• Support for various conditioning modes