v1.0.0

Kedreamix/Linly-Talker-Streamv1.0.0Mar 17, 2024by FrankLeeeee

AI Summary

This release introduces Open-Sora, an open-source initiative dedicated to democratizing high-quality video generation. It provides complete source code for data processing, training, and inference pipelines, optimized with ColossalAI and flash attention. Pre-trained model weights are also available for 16x256x256 and 16x512x512 resolutions.

Key Highlights

Open-source initiative for high-quality video generation
Complete source code for data processing and sora-like pipelines
System optimizations using ColossalAI and flash attention
Pre-trained model weights for 16x256x256 and 16x512x512 resolutions

New Features

Source code for data processing
Source code for sora-like video generation training and inference pipeline
System optimization powered by ColossalAI and flash attention
Model weights trained for 16x256x256 and 16x512x512 resolutions

Full Release Notes

We present Open-Sora, an initiative dedicated to efficiently produce high-quality video and make the model, tools and contents accessible to all. By embracing open-source principles, Open-Sora not only democratizes access to advanced video generation techniques, but also offers a streamlined and user-friendly platform that simplifies the complexities of video production. With Open-Sora, we aim to inspire innovation, creativity, and inclusivity in the realm of content creation.

In this release, we have included
- source code for data processing
- source code for sora-like video generation training and inference pipeline
- system optimization powered by ColossalAI and flash attention
- model weights trained for 16×256×256 and 16×512x512 resolutions, available on Hugging Face