Wan2.2-Animate-14B-GGUF is a quantized version of the Wan2.2-Animate-14B video-to-video model provided in GGUF format. It is hosted on Hugging Face by QuantStack and is intended for use in video-to-video generation tasks.
The repository supplies the main model files along with guidance on complementary components including the Umt5-xxl text encoder and Wan2.1_VAE. It lists multiple quantization levels that allow users to trade off between model size and performance on available hardware. Available options range from 2-bit Q2_K at 6.46 GB through 8-bit Q8_0 at 18.7 GB. An example workflow for integration is referenced.
The model is delivered as downloadable GGUF files placed in the ComfyUI/models/unet directory. It is designed for operation with the ComfyUI-GGUF custom node. The original model originates from Wan-AI/Wan2.2-Animate-14B and retains all original licensing terms and usage restrictions. The repository itself is released under the Apache-2.0 license. It has recorded 116000 downloads in the past month.
The tool belongs to the class of video-to-video models and supports English and Chinese. No pricing information is stated because the files are offered for direct download.
Wan2.2 Animate 14B sits in PulseGate's Video generation category. It focuses on running high-quality video generation and video-to-video models locally with reduced memory using GGUF quantization. Wan2.2 Animate 14B is an open-source project aimed at developers. Wan2.2 Animate 14B is open source under the Apache-2.0 license. It runs on the web and API.
QuantStack builds and maintains Wan2.2 Animate 14B, and it first shipped in 2024. Development happens publicly on GitHub with 3.8k stars. Among its 4 catalogued features are video generation, video-to-video, and GGUF quantization.
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