[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-78612":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":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":23,"hasPages":23,"topics":25,"createdAt":10,"pushedAt":10,"updatedAt":31,"readmeContent":32,"aiSummary":33,"trendingCount":16,"starSnapshotCount":16,"syncStatus":34,"lastSyncTime":35,"discoverSource":36},78612,"lyra","nv-tlabs\u002Flyra","nv-tlabs","Project Lyra: Open Generative 3D World Models","https:\u002F\u002Fresearch.nvidia.com\u002Flabs\u002Fsil\u002Fprojects\u002Flyra2\u002F",null,"Python",2081,225,39,15,0,11,30,54,33,29.06,"Apache License 2.0",false,"main",[26,27,28,29,30],"3d","diffusion","gaussians","video","world-model","2026-06-12 02:03:47","\u003Cp align=\"center\">\n\u003Ch1 align=\"center\">Project Lyra: Open Generative 3D World Models\u003C\u002Fh1>\n\u003Ch3 align=\"center\">NVIDIA Spatial Intelligence Lab\u003C\u002Fh3>\n\u003C\u002Fp>\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002Ff47c24c0-453e-4134-84f1-80c56613f4af\n\n\u003C!-- \u003Cp align=\"center\">\n  \u003Cimg src=\"https:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F12d44362-8b7f-4952-9488-0e45cf759b57\" alt=\"teaser\"\u002F>\n\u003C\u002Fp> -->\n\n## 🫨 News\n\n- ```2026-04-15```: 👋 We released [Lyra-2.0](https:\u002F\u002Fgithub.com\u002Fnv-tlabs\u002Flyra\u002Ftree\u002Fmain\u002FLyra-2). Explorable generative 3D worlds with long-horizon, 3D-consistent generation.\n- ```2025-09-23```: 👋 We released [Lyra-1.0](https:\u002F\u002Fgithub.com\u002Fnv-tlabs\u002Flyra\u002Ftree\u002Fmain\u002FLyra-1). Feed-forward 3D and 4D scene generation from a single image\u002Fvideo via video diffusion model self-distillation.\n\n## 📝 Overview\n\n**Project Lyra** is a series of open generative 3D world models developed at NVIDIA.\n\nThis repository provides the official implementations of Lyra 1.0 and Lyra 2.0.\n\n| Version | 📄 Paper | 🌐 Project Page | 🤗 Model | 💻 Code |\n|---------|----------|-----------------|----------|---------|\n| Lyra 1.0 | [![arXiv](https:\u002F\u002Fimg.shields.io\u002Fstatic\u002Fv1?label=&message=arXiv&color=red&logo=arxiv)](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.19296) | [![Page](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Project%20Page-00bfff)](https:\u002F\u002Fresearch.nvidia.com\u002Flabs\u002Ftoronto-ai\u002Flyra\u002F) | [![HuggingFace](https:\u002F\u002Fimg.shields.io\u002Fstatic\u002Fv1?label=&message=HuggingFace&color=yellow)](https:\u002F\u002Fhuggingface.co\u002Fnvidia\u002FLyra) | [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Lyra--1-blue)](Lyra-1\u002F) |\n| Lyra 2.0 |  | [![Page](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Project%20Page-00bfff)](https:\u002F\u002Fresearch.nvidia.com\u002Flabs\u002Fsil\u002Fprojects\u002Flyra2\u002F) | [![HuggingFace](https:\u002F\u002Fimg.shields.io\u002Fstatic\u002Fv1?label=&message=HuggingFace&color=yellow)](https:\u002F\u002Fhuggingface.co\u002Fnvidia\u002FLyra-2.0) | [![Code](https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F-Lyra--2-blue)](Lyra-2\u002F) |\n\n## License\n\nLyra source code is released under the [Apache 2.0 License](LICENSE). Please refer to [Lyra-1](Lyra-1\u002F) and [Lyra-2](Lyra-2\u002F) for their respective model licenses.\n\n## Citation\nLyra 2.0 is temporarily anonymized. Please Google for citation. Thanks!\n\n```bibtex\n@inproceedings{bahmani2026lyra,\n  title={Lyra: Generative 3D Scene Reconstruction via Video Diffusion Model Self-Distillation},\n  author={Bahmani, Sherwin and Shen, Tianchang and Ren, Jiawei and Huang, Jiahui and Jiang, Yifeng and \n          Turki, Haithem and Tagliasacchi, Andrea and Lindell, David B. and Gojcic, Zan and Fidler, Sanja and \n          Ling, Huan and Gao, Jun and Ren, Xuanchi},\n  booktitle={International Conference on Learning Representations (ICLR)},\n  year={2026}\n}\n```\n","Project Lyra 是 NVIDIA 开发的一系列开放生成式3D世界模型。该项目的核心功能是通过视频扩散模型自蒸馏技术，从单张图像或视频生成可探索的3D场景，并支持长时间范围内的3D一致性生成。Lyra 1.0和Lyra 2.0版本分别提供了前馈3D、4D场景生成以及更复杂的3D世界构建能力。项目使用Python语言编写，具有良好的开源社区支持。适用于需要高质量3D内容生成的场景，如虚拟现实、游戏开发、建筑设计等领域。",2,"2026-06-11 03:57:01","high_star"]