---
title: vqvae
canonical_url: "https://www.modelscope.cn/models/AI-ModelScope/vqvae"
md_url: "https://www.modelscope.cn/models/AI-ModelScope/vqvae.md"
repository: AI-ModelScope/vqvae
last_updated: 2024-03-18
license: mit
model_type:
  - VQVAE
architectures:
  - VQVAE
parameters: 22.2M
tensor_type:
  - F32
  - I64
library_name:
  - safetensors
  - pytorch
frameworks:
  - pytorch
downloads: 4329
stars: 0
---

# vqvae

> vqvae - AI-ModelScope 在 ModelScope 开源的模型。This repository is a clone of the VideoGPT in order to convert the VQ-VAE model to the Hugging Face format for easier model loading.

AI-ModelScope/vqvae 是 ModelScope 魔搭社区上的 22.2M 参数机器学习模型，采用 mit 许可。

- **Repository**: AI-ModelScope/vqvae
- **License**: mit
- **Parameters**: 22.2M
- **Downloads**: 4329
- **Stars**: 0
- **Last updated**: 2024-03-18

Source: https://www.modelscope.cn/models/AI-ModelScope/vqvae

---

# VQVAE

This repository is a clone of the [VideoGPT](https://github.com/wilson1yan/VideoGPT/tree/master) in order to convert the VQ-VAE model to the Hugging Face format for easier model loading.

Paper: [VideoGPT: Video Generation using VQ-VAE and Transformers](https://arxiv.org/abs/2104.10157)

## License

We follow the MIT license distributed by the [VideoGPT](https://github.com/wilson1yan/VideoGPT/tree/master) project.