Wan2.1 I2V 14B 480P is an image-to-video diffusion model hosted on Hugging Face. It generates short video clips from a starting image combined with a text prompt and is provided as an open-source model under the Apache 2.0 license.
The model carries 14 billion parameters and produces 480p output. It is tagged for i2v and video-generation tasks and supports both English and Chinese. Integration with the Diffusers library allows users to load the pipeline, supply an image and prompt, and export the resulting frames as video. Example code demonstrates installation via pip, use of bfloat16 precision on CUDA devices, and conversion of output to an mp4 file.
Distribution occurs through the Hugging Face repository, where the model card supplies instructions for local execution and references to inference providers. The repository includes Safetensors format files and supports deployment options such as notebooks or local applications. No pricing information appears for the model itself, which remains freely available for download and use under its stated license.
Wan2.1 I2V 14B 480P is an Image to video project. It focuses on converting static images into short coherent videos using text guidance. It is built as an open-source project for AI developers. The project is open source (Apache-2.0). Wan2.1 I2V 14B 480P is available on the web, the command line, and API.
Behind Wan2.1 I2V 14B 480P is Wan-AI, and it first shipped in 2025. The project is developed in the open on GitHub with 16.6k stars. Key capabilities include image to Video, Text Conditioning, and Diffusers Support.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Closest matches by what these projects do