Wan2.2 I2V A14B is an image-to-video diffusion model hosted on Hugging Face. It accepts a starting image and a text prompt to produce a short video clip.
The model is provided in Diffusers format and uses the WanImageToVideoPipeline. Installation proceeds through the command pip install -U diffusers transformers accelerate. Code examples load the pipeline with bfloat16 precision, map it to a CUDA device, pass an image loaded via load_image and a prompt string such as "A man with short gray hair plays a red electric guitar," then retrieve the generated frames and export them with export_to_video. The repository supplies the model in Safetensors format and carries an Apache-2.0 license. An arXiv paper numbered 2503.20314 is referenced in the repository metadata.
It is delivered as a downloadable model card on the Hugging Face platform. The repository lists support for English and Chinese. Users run it locally or on compatible inference providers after installing the listed libraries. The page indicates 11.7k likes and 280 followers for the Wan-AI organization.
Wan2.2 I2V A14B is a Video generation product. It focuses on generating coherent video sequences from static images using large-scale diffusion models. Wan2.2 I2V A14B is an open-source project aimed at AI researchers and video creators. The project is open source (Apache-2.0). Wan2.2 I2V A14B is available on the web, the command line, and API.
It is developed by Wan-AI, and the product first shipped in 2025. The project is developed in the open on GitHub with 16.8k stars. PulseGate's similarity index places it among 6 comparable tools. Among its 3 catalogued features are image-to-video, diffusers integration, and high-resolution output.
Latest indexed changes and source events
Wan-AI/Wan2.2-I2V-A14B-Diffusers verified by the PulseGate indexer
Other apps tracked under the same category.
