Skip to content

Allegro-TI2V

Open text-and-image to video generation

Visit official site

What Allegro-TI2V is

The model is described as a text-image-to-video generation model that creates video content from a user prompt and an image, with open weights and code.

What you can do with it

  • Generate video from a prompt and a first frame
  • Continue a clip from first and last frames
  • Produce varied scenes from people to animals

Getting started

  • Download the Allegro code from GitHub
  • Install dependencies with Python 3.10+ and PyTorch 2.4+
  • Fetch the model weights from Hugging Face
  • Run the command line tool with a prompt and image path

Pros & cons

✓ What we like

  • Weights and code are open under Apache 2.0
  • FP32, BF16 and FP16 precision are supported

! What to watch out for

  • Python 3.10 or newer and PyTorch 2.4 or newer are listed
  • EMA-VFI interpolation is used to reach 30FPS

FAQ

What license applies?

The material states the model follows the Apache 2.0 license.

What are the output specs?

The feature list names 720x1280, 15FPS, 6 seconds and interpolation to 30FPS.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review