[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-71935":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":17,"stars30d":18,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":19,"rankGlobal":10,"rankLanguage":10,"license":20,"archived":21,"fork":21,"defaultBranch":22,"hasWiki":21,"hasPages":23,"topics":24,"createdAt":10,"pushedAt":10,"updatedAt":25,"readmeContent":26,"aiSummary":27,"trendingCount":16,"starSnapshotCount":16,"syncStatus":28,"lastSyncTime":29,"discoverSource":30},71935,"Open-Sora-Plan","PKU-YuanGroup\u002FOpen-Sora-Plan","PKU-YuanGroup","This project aim to reproduce Sora (Open AI T2V model), we wish the open source community contribute to this project.","",null,"Python",12160,1066,149,252,0,3,6,71.18,"MIT License",false,"main",true,[],"2026-06-12 04:01:02","\n\n\u003Ch1 align=\"left\"> \u003Ca href=\"\">Open-Sora Plan\u003C\u002Fa>\u003C\u002Fh1>\n\nThis project aims to create a simple and scalable repo, to reproduce [Sora](https:\u002F\u002Fopenai.com\u002Fsora) (OpenAI, but we prefer to call it \"ClosedAI\" ). \n\n本项目希望通过开源社区的力量复现Sora，由北大-兔展AIGC联合实验室共同发起，来自兔展、华为、鹏城实验室和开源社区伙伴均有深度贡献力量。\n\n当前V1.5版本**完全基于华为昇腾训练（昇腾纯血版）**，欢迎Pull Request和使用！\n\n我们正在快速迭代新版本，欢迎更多合作者或算法工程师加入，[算法工程师招聘-兔展智能.pdf](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Ffiles\u002F19107972\u002F-.pdf)\n\n\u003Ch5 align=\"left\">\n\n[![arXiv](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FArxiv-Open--Sora%20Plan-b31b1b.svg?logo=arXiv)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.00131)\n[![arXiv](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FArxiv-Helios-b31b1b.svg?logo=arXiv)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2603.04379)\n[![arXiv](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FArxiv-WF--VAE-b31b1b.svg?logo=arXiv)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.17459)\n[![License](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLicense-Apache-yellow)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fblob\u002Fmain\u002FLICENSE)  \u003Cbr>\n[![slack badge](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-join-blueviolet?logo=discord&amp)](https:\u002F\u002Fdiscord.gg\u002FDFZg5678)\n[![WeChat badge](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F微信-加入-green?logo=wechat&amp)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fissues\u002F53#issuecomment-1987226516)\n[![Twitter](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Twitter@LinBin46984-black?logo=twitter&logoColor=1D9BF0)](https:\u002F\u002Fx.com\u002FLinBin46984\u002Fstatus\u002F1795018003345510687) \n[![Modelers](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%E9%AD%94%E4%B9%90-%E6%A8%A1%E5%9E%8B%E4%BD%93%E9%AA%8C-blue)](https:\u002F\u002Fmodelers.cn\u002Fspaces\u002FMindSpore-Lab\u002FOpen_Sora_Plan) \u003Cbr>\n[![GitHub repo stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?style=flat&logo=github&logoColor=whitesmoke&label=Stars)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fstargazers) \n[![GitHub repo forks](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fforks\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?style=flat&logo=github&logoColor=whitesmoke&label=Forks)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fnetwork) \n[![GitHub repo watchers](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fwatchers\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?style=flat&logo=github&logoColor=whitesmoke&label=Watchers)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fwatchers) \n[![GitHub repo size](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Frepo-size\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?style=flat&logo=github&logoColor=whitesmoke&label=Repo%20Size)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Farchive\u002Frefs\u002Fheads\u002Fmain.zip) \u003Cbr>\n[![GitHub repo contributors](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcontributors-anon\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?style=flat&label=Contributors)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fgraphs\u002Fcontributors) \n[![GitHub Commit](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fcommit-activity\u002Fm\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?label=Commit)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fcommits\u002Fmain\u002F)\n[![Pr](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-pr-closed-raw\u002FPKU-YuanGroup\u002FOpen-Sora-Plan.svg?label=Merged+PRs&color=green)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fpulls)\n[![GitHub issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?color=critical&label=Issues)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FVideo-LLaVA\u002Fissues?q=is%3Aopen+is%3Aissue)\n[![GitHub closed issues](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fissues-closed\u002FPKU-YuanGroup\u002FOpen-Sora-Plan?color=success&label=Issues)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FVideo-LLaVA\u002Fissues?q=is%3Aissue+is%3Aclosed)\n\u003C\u002Fh5>\n\u003Ca href=\"https:\u002F\u002Ftrendshift.io\u002Frepositories\u002F8280\" target=\"_blank\">\u003Cimg src=\"https:\u002F\u002Ftrendshift.io\u002Fapi\u002Fbadge\u002Frepositories\u002F8280\" alt=\"PKU-YuanGroup%2FOpen-Sora-Plan | Trendshift\" style=\"width: 250px; height: 55px;\" width=\"250\" height=\"55\"\u002F>\u003C\u002Fa>\n\u003Ch5 align=\"left\"> If you like our project, please give us a star ⭐ on GitHub for latest update.  \u003C\u002Fh2>\n\n\n# 📣 News\n\n* **[2026.03.08]** 👋👋👋 We introduce [Helios](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FHelios), a breakthrough video generation model that achieves minute-scale, high-quality video synthesis at **19.5 FPS on a single H100** GPU — without relying on conventional long video anti-drifting strategies or standard video acceleration techniques. Welcome to check [Technical Report](https:\u002F\u002Fhuggingface.co\u002Fpapers\u002F2603.04379)!\n* **[2025.06.05]** 🔥🔥🔥 We release version 1.5.0, our most powerful model! By introducing a **higher-compression WFVAE** and an improved sparse DiT architecture, **SUV**, we achieve performance **comparable to HunyuanVideo (Open-Source)** using an 8B-scale model and 40 million video samples. Version 1.5.0 is **fully trained and inferred on Ascend 910-series accelerators**; Please check the [mindspeed_mmdit](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Ftree\u002Fmindspeed_mmdit) branch for our new code and [Report-v1.5.0.md](docs\u002FReport-v1.5.0.md) for our report. The GPU version is coming soon. \n* **[2024.12.03]** ⚡️ We released our [arxiv paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.00131) and WF-VAE [paper](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.17459) for v1.3. The next more powerful version is coming soon.\n* **[2024.10.16]** 🎉 We released version 1.3.0, featuring: **WFVAE**, **prompt refiner**, **data filtering strategy**, **sparse attention**, and **bucket training strategy**. We also support 93x480p within **24G VRAM**. More details can be found at our latest [report](docs\u002FReport-v1.3.0.md).\n* **[2024.08.13]** 🎉 We are launching Open-Sora Plan v1.2.0 **I2V** model, which is based on Open-Sora Plan v1.2.0. The current version supports image-to-video generation and transition generation (the starting and ending frames conditions for video generation). Check out the Image-to-Video section in this [report](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fblob\u002Fmain\u002Fdocs\u002FReport-v1.2.0.md#training-image-to-video-diffusion-model).\n* **[2024.07.24]** 🔥🔥🔥 v1.2.0 is here! Utilizing a 3D full attention architecture instead of 2+1D. We released a true 3D video diffusion model trained on 4s 720p. Check out our latest [report](docs\u002FReport-v1.2.0.md).\n* **[2024.05.27]** 🎉 We are launching Open-Sora Plan v1.1.0, which significantly improves video quality and length, and is fully open source! Please check out our latest [report](docs\u002FReport-v1.1.0.md). Thanks to [ShareGPT4Video's](https:\u002F\u002Fsharegpt4video.github.io\u002F) capability to annotate long videos.\n* **[2024.04.09]** 🤝 Excited to share our latest exploration on metamorphic time-lapse video generation: [MagicTime](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FMagicTime), which learns real-world physics knowledge from time-lapse videos.\n* **[2024.04.07]** 🎉🎉🎉 Today, we are thrilled to present Open-Sora-Plan v1.0.0, which significantly enhances video generation quality and text control capabilities. See our [report](docs\u002FReport-v1.0.0.md). Thanks to HUAWEI NPU for supporting us.\n* **[2024.03.27]** 🚀🚀🚀 We release the report of [VideoCausalVAE](docs\u002FCausalVideoVAE.md), which supports both images and videos. We present our reconstructed video in this demonstration as follows. The text-to-video model is on the way.\n* **[2024.03.01]** 🤗 We launched a plan to reproduce Sora, called Open-Sora Plan! Welcome to **watch** 👀 this repository for the latest updates.\n\n# 😍 Gallery\n\nText-to-Video Generation of Open-Sora Plan v1.5.0.\n### Youtube:\n[![Demo Video of Open-Sora Plan V1.5.0](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F130bbba2-3ded-4092-92ef-b65b673cb1a6)](https:\u002F\u002Fyoutu.be\u002FIiWTdx2EHCY)\n### Bilibili:\n[![Demo Video of Open-Sora Plan V1.5.0](https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F130bbba2-3ded-4092-92ef-b65b673cb1a6)](https:\u002F\u002Fwww.bilibili.com\u002Fvideo\u002FBV1X77tzxE3b\u002F)\n\n# 😮 Highlights\n\nOpen-Sora Plan shows excellent performance in video generation.\n\n### 🔥 WFVAE with higher performance and compression\n- With an 8×8×8 downsampling rate, but achieves higher PSNR than the VAE used in Wan2.1. Lowers the training cost for the DiT built upon it.\n\n### 🚀 More powerful sparse dit\n- The more powerful sparse attention architecture, SUV, achieves performance close to dense DiT while providing over a 35% speedup.\n\n\u003Cp align=\"center\">\n    \u003Cimg src=\"https:\u002F\u002Fs21.ax1x.com\u002F2024\u002F07\u002F22\u002Fpk7cob8.png\" width=\"650\" style=\"margin-bottom: 0.2;\"\u002F>\n\u003Cp>\n\n# 🐳 Resource\n\n| Version | Architecture |  Diffusion Model | CausalVideoVAE | Data | Prompt Refiner |\n|:---|:---|:---|:---|:---|:---|\n| v1.5.0 | SUV (Skiparse 3D) | [121x576x1024](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.5.0\u002Fblob\u002Fmain\u002FMindSpeed\u002Fmodel_ema.pt)[5] | [Anysize_8x8x8_32dim](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.5.0\u002Fblob\u002Fmain\u002FMindSpeed\u002Fwfvae_888_dim32.ckpt) | - | - |\n| v1.3.0 [4] | Skiparse 3D | [Anysize in 93x640x640](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.3.0\u002Ftree\u002Fmain\u002Fany93x640x640)[3], [Anysize in 93x640x640_i2v](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.3.0\u002Ftree\u002Fmain\u002Fany93x640x640_i2v)[3] | [Anysize](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.3.0\u002Ftree\u002Fmain\u002Fvae)| [prompt_refiner](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FLanguageBind\u002FOpen-Sora-Plan-v1.3.0\u002Ftree\u002Fmain\u002Fprompt_refiner) | [checkpoint](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.3.0\u002Ftree\u002Fmain\u002Fprompt_refiner)| |\n| v1.2.0 | Dense 3D | [93x720p](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0\u002Ftree\u002Fmain\u002F93x720p), [29x720p](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0\u002Ftree\u002Fmain\u002F29x720p)[1], [93x480p](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0\u002Ftree\u002Fmain\u002F93x480p)[1,2], [29x480p](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0\u002Ftree\u002Fmain\u002F29x480p), [1x480p](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0\u002Ftree\u002Fmain\u002F1x480p), [93x480p_i2v](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0\u002Ftree\u002Fmain\u002F93x480p_i2v) | [Anysize](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0\u002Ftree\u002Fmain\u002Fvae)| [Annotations](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FLanguageBind\u002FOpen-Sora-Plan-v1.2.0) | - |\n| v1.1.0 | 2+1D | [221x512x512](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.1.0\u002Ftree\u002Fmain\u002F221x512x512), [65x512x512](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.1.0\u002Ftree\u002Fmain\u002F65x512x512) |[Anysize](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.1.0\u002Ftree\u002Fmain\u002Fvae) |[Data and Annotations](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FLanguageBind\u002FOpen-Sora-Plan-v1.1.0)| - |\n| v1.0.0 | 2+1D | [65x512x512](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.0.0\u002Ftree\u002Fmain\u002F65x512x512), [65x256x256](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.0.0\u002Ftree\u002Fmain\u002F65x256x256), [17x256x256](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.0.0\u002Ftree\u002Fmain\u002F17x256x256) | [Anysize](https:\u002F\u002Fhuggingface.co\u002FLanguageBind\u002FOpen-Sora-Plan-v1.0.0\u002Ftree\u002Fmain\u002Fvae) | [Data and Annotations](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FLanguageBind\u002FOpen-Sora-Plan-v1.0.0)| - |\n\n> [1] Please note that the weights for v1.2.0 29×720p and 93×480p were trained on Panda70M and have not undergone final high-quality data fine-tuning, so they may produce watermarks.\n\n> [2] We fine-tuned 3.5k steps from 93×720p to get 93×480p for community research use.\n\n> [3] The model is trained arbitrarily on stride=32. So keep the resolution of the inference a multiple of 32. Frames need to be 4n+1, e.g. 93, 77, 61, 45, 29, 1 (image).\n\n> [4] Model weights are also available at [OpenMind](https:\u002F\u002Fmodelers.cn\u002Fmodels\u002Flinbin\u002FOpen-Sora-Plan-v1.3.0) and [WiseModel](https:\u002F\u002Fwisemodel.cn\u002Fmodels\u002FPKU-YUAN\u002FOpen-Sora-Plan-v1.3.0).\n\n> [5] The current model weights are only compatible with the NPU + MindSpeed-MM framework. Model weights are also available at and [modelers](https:\u002F\u002Fmodelers.cn\u002Fmodels\u002FPKU-YUAN-Group\u002FOpen-Sora-Plan-v1.5.0\u002Ftree\u002Fmain\u002FMindSpeed).\n\n> [!Warning]\n>\n> \u003Cdiv align=\"left\">\n> \u003Cb>\n> 🚨 For version 1.2.0, we no longer support 2+1D models.\n> \u003C\u002Fb>\n> \u003C\u002Fdiv>\n\n# ⚙️ How to start\n\n### GPU\ncoming soon...\n### NPU\nPlease check out the **[mindspeed_mmdit](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Ftree\u002Fmindspeed_mmdit)** branch and follow the README.md for configuration.\n\n# 📖 Technical report\nPlease check [Report-v1.5.0.md](docs\u002FReport-v1.5.0.md).\n\n# 💡 How to Contribute\nWe greatly appreciate your contributions to the Open-Sora Plan open-source community and helping us make it even better than it is now!\n\nFor more details, please refer to the [Contribution Guidelines](docs\u002FContribution_Guidelines.md)\n\n# 👍 Acknowledgement and Related Work\n* [Allegro](https:\u002F\u002Fgithub.com\u002Frhymes-ai\u002FAllegro): Allegro is a powerful text-to-video model that generates high-quality videos up to 6 seconds at 15 FPS and 720p resolution from simple text input based on our Open-Sora Plan. The significance of open-source is becoming increasingly tangible.\n* [Latte](https:\u002F\u002Fgithub.com\u002FVchitect\u002FLatte): It is a wonderful 2+1D video generation model.\n* [PixArt-alpha](https:\u002F\u002Fgithub.com\u002FPixArt-alpha\u002FPixArt-alpha): Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis.\n* [ShareGPT4Video](https:\u002F\u002Fgithub.com\u002FInternLM\u002FInternLM-XComposer\u002Ftree\u002Fmain\u002Fprojects\u002FShareGPT4Video): Improving Video Understanding and Generation with Better Captions.\n* [VideoGPT](https:\u002F\u002Fgithub.com\u002Fwilson1yan\u002FVideoGPT): Video Generation using VQ-VAE and Transformers.\n* [DiT](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FDiT): Scalable Diffusion Models with Transformers.\n* [FiT](https:\u002F\u002Fgithub.com\u002Fwhlzy\u002FFiT): Flexible Vision Transformer for Diffusion Model.\n* [Positional Interpolation](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.15595): Extending Context Window of Large Language Models via Positional Interpolation.\n\n\n# 🔒 License\n* See [LICENSE](LICENSE) for details.\n\n## ✨ Star History\n\n[![Star History](https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=PKU-YuanGroup\u002FOpen-Sora-Plan)](https:\u002F\u002Fstar-history.com\u002F#PKU-YuanGroup\u002FOpen-Sora-Plan&Date)\n\n\n# ✏️ Citing\n\n\n```bibtex\n@article{lin2024open,\n  title={Open-Sora Plan: Open-Source Large Video Generation Model},\n  author={Lin, Bin and Ge, Yunyang and Cheng, Xinhua and Li, Zongjian and Zhu, Bin and Wang, Shaodong and He, Xianyi and Ye, Yang and Yuan, Shenghai and Chen, Liuhan and others},\n  journal={arXiv preprint arXiv:2412.00131},\n  year={2024}\n}\n```\n```bibtex\n@article{helios,\n  title={Helios: Real Real-Time Long Video Generation Model},\n  author={Yuan, Shenghai and Yin, Yuanyang and Li, Zongjian and Huang, Xinwei and Yang, Xiao and Yuan, Li},\n  journal={arXiv preprint arXiv:2603.04379},\n  year={2026}\n}\n```\n```bibtex\n@article{li2024wf,\n  title={WF-VAE: Enhancing Video VAE by Wavelet-Driven Energy Flow for Latent Video Diffusion Model},\n  author={Li, Zongjian and Lin, Bin and Ye, Yang and Chen, Liuhan and Cheng, Xinhua and Yuan, Shenghai and Yuan, Li},\n  journal={arXiv preprint arXiv:2411.17459},\n  year={2024}\n}\n```\n\n# 🤝 Community contributors\n\n\u003Ca href=\"https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpen-Sora-Plan\u002Fgraphs\u002Fcontributors\">\n  \u003Cimg src=\"https:\u002F\u002Fcontrib.rocks\u002Fimage?repo=PKU-YuanGroup\u002FOpen-Sora-Plan\" \u002F>\n\u003C\u002Fa>\n\n","Open-Sora Plan 项目旨在通过开源社区的力量复现 OpenAI 的 Sora 模型。该项目采用 Python 编写，当前版本基于华为昇腾平台进行训练，具备高度的可扩展性和简洁性。其核心功能包括文本到视频生成，支持社区贡献和快速迭代。适用于需要高质量文本转视频的应用场景，如内容创作、广告制作等。项目已获得广泛支持，拥有活跃的开发者社区。",2,"2026-06-11 03:39:34","high_star"]