v1.3

zaidmukaddam/scirav1.3Feb 21, 2025by zhengzangw

AI Summary

Open-Sora V1.3 introduces architectural improvements and enhanced training data to boost video generation quality, while maintaining support for image-to-video and video extension tasks.

Key Highlights

Upgraded Video Compression
Enhanced ST-DiT Architecture
Multi-Stage Training with high-quality data
Advanced model conditioning and generation quality

New Features

Upgraded video compression
Enhanced ST-DiT architecture
Multi-stage training with high-quality data
Advanced model conditioning
Expanded resolution and frame length support
Support for image-to-video generation
Support for video extension

Full Release Notes

We are excited to introduce Open-Sora V1.3.0, bringing improvements in model architecture and generation quality.

Open-Sora V1.3.0 Release 🎉

### 🚀 Key Highlights:
📦 Upgraded Video Compression
🌀 Enhanced ST-DiT Architecture
🎞️ More High-Quality Data & Multi-Stage Training
✨ Advanced Model Conditioning and Generation Quality

Open-Sora 1.3 continues to support image-to-video generation and video extension with expanded resolution and frame length support.

🔗 Check out our [technical report](https://github.com/hpcaitech/Open-Sora/blob/main/docs/report_04.md) and [gallery](https://hpcaitech.github.io/Open-Sora/) to explore more!