Skip to main content
Rase AI Intelligence Newsletter

Creative AI Model or AI service

Wan2.1 releases models for generating video from text and images

The Wan team published downloadable video generation models that create clips from text prompts or animate a supplied image. The original release includes multiple checkpoints. [1][2]

Model access
Downloadable text-to-video and image-to-video checkpoints, with project inference code. [1][2]
License
The official repository lists Apache License 2.0 for these models, subject to its stated terms. [1]

What the model does

On February 25, 2025, the project released Wan2.1 weights and inference code. Its initial choices included a smaller text-to-video model and larger text-to-video or image-to-video variants. [1][2]

Why it matters

A developer can choose a model according to whether the starting point is a written brief or a still image. That distinction affects how much of an existing design a clip is expected to retain. [1][2]

Who can use it

Creative teams can evaluate the downloadable checkpoints for storyboards or concept clips. Researchers can compare how the text-to-video and image-to-video models respond to the same scene description. [1][2]

Key dates

  1. Original Wan2.1 weights and code released [1]

Ways to use it

Use an original or authorized image to test an animation of a product sketch. Compare how consistently the clip preserves the main design and whether it follows the requested movement. [1]

Model limitations

Memory needs and supported resolutions depend on the variant. Results can change between generated clips, so review motion and visual consistency before reuse. [1][2]

Sources and verification

Catalog evidence checked How we verify records

  1. Primary source Wan team’s dated Wan2.1 release, checkpoints and license
  2. Primary source Wan-AI/Wan2.1-T2V-14B

Related profiles

Creative AI

Seedance 2.5 (4K API and Flash)

ByteDance (Volcano Engine / BytePlus)

These tools generate video through developer services. The cited sources describe different variants and routes of access.

Medium confidenceEvidence checked

Creative AI

LongCat-Video-Avatar 1.6

Meituan (LongCat team)

It generates video of an avatar from supported inputs in the documented project.

High confidenceEvidence checked

Creative AI

MiniMax H3-Image

MiniMax

It generates images from prompts through the model architecture described by MiniMax.

High confidenceEvidence checked