# PKU-YuanGroup/Open-Sora-Plan

This project aim to reproduce Sora (Open AI T2V model), we wish the open source community contribute to this project.

Repository: https://github.com/PKU-YuanGroup/Open-Sora-Plan
Canonical: https://ross.abutalabs.com/products/open-sora-plan
Language: Python
License: MIT
License Family: permissive
Last push: 2026-03-08T12:48:46+00:00

## Health v2 (maintenance only)
Score: 51/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 71, release rhythm 16, longevity 66
- inputs: {"age_days": 925, "days_push": 178, "days_rel": 454, "gap_med": 115.5, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 12155, forks 1060 (observed 2026-08-28T04:10:51.856007+00:00)

## What it is
Open-Sora Plan is an open-source effort to reproduce OpenAI's Sora text-to-video model, providing training and inference code for video generation models including a WF-VAE and DiT-based architectures. It is developed by the PKU-Yuan Group with community contributions and supports training on Ascend NPUs.

## Use cases
- generate videos from text prompts
- train a text-to-video diffusion model
- reproduce Sora-style video generation
- run open-source video generation inference
- fine-tune a video generation model on custom data
- research video generation architectures like VAE and DiT

## When to choose
- you need an open-source, self-hosted text-to-video model you can inspect and modify
- you want to train or fine-tune video generation models on your own hardware including Ascend NPUs
- you are doing research on video diffusion architectures

## When to avoid
- you need a polished production-ready video generation service with minimal setup
- you lack significant GPU/NPU resources, since training is compute-intensive
- you only need simple image generation rather than video

## Facets
- artifact type: library
- maturity: active
- function: deep-learning, machine-learning, video-processing, llm-training
- domain: deep-learning, artificial-intelligence, image-processing
- platform: python
- tags: text-to-video, video-generation, diffusion-models, sora-reproduction, open-source-model, vae, transformers, video, gpu, linux

## Member repositories
- PKU-YuanGroup/Open-Sora-Plan (main) score 51

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:51.856007+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:14:53.754380+00:00, confidence not recorded.
  - readme: https://github.com/PKU-YuanGroup/Open-Sora-Plan (fetched 2026-08-28T04:10:51.856007+00:00, sha 5ffa75d546e6)
- Data as of 2026-08-30T08:39:29.467469+00:00.