Skip to content

HunyuanVideo-I2V

Image-to-video generation from Tencent

Visit official site

What hunyuanvideo-i2v is

The model is described as a Tencent open-source image-to-video model based on the HunyuanVideo architecture that folds reference images into generation.

What you can do with it

  • Animate a still figure into a short motion clip
  • Train a LoRA effect for a special visual result
  • Bring a scenery photo to life as moving footage

Key capabilities

  • 720p video output is supported
  • Pretrained weights and inference code are released
  • Multi-GPU parallel inference speeds up generation

Pros & cons

✓ What we like

  • Custom LoRA training enables distinct visual effects
  • Multiple operating systems and hardware setups are compatible

! What to watch out for

  • The repo and pretrained weights must be downloaded first
  • A reference image and a text prompt are needed before inference

FAQ

What resolution is supported?

The material lists high-resolution 720p video generation.

Is it open source?

The material describes it as an open-source image-to-video model.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review