[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-72400":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":35,"readmeContent":36,"aiSummary":37,"trendingCount":16,"starSnapshotCount":16,"syncStatus":38,"lastSyncTime":39,"discoverSource":40},72400,"HunyuanWorld-1.0","Tencent-Hunyuan\u002FHunyuanWorld-1.0","Tencent-Hunyuan","Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels with Hunyuan3D World Model","https:\u002F\u002F3d.hunyuan.tencent.com\u002FsceneTo3D",null,"Python",2844,257,30,28,0,6,13,36,18,81.33,"Other",false,"main",[26,27,28,29,30,31,32,33,34],"3d","3d-generation","aigc","hunyuan3d","image-to-3d","scene-generation","text-to-3d","world-model","world-models","2026-06-12 04:01:05","[中文阅读](README_zh_cn.md)\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fteaser.png\">\n\u003C\u002Fp>\n\n\u003Cdiv align=\"center\">\n  \u003Ca href=https:\u002F\u002F3d.hunyuan.tencent.com\u002FsceneTo3D target=\"_blank\">\u003Cimg src=https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FOfficial%20Site-333399.svg?logo=homepage height=22px>\u003C\u002Fa>\n  \u003Ca href=https:\u002F\u002Fhuggingface.co\u002Ftencent\u002FHunyuanWorld-1 target=\"_blank\">\u003Cimg src=https:\u002F\u002Fimg.shields.io\u002Fbadge\u002F%F0%9F%A4%97%20Models-d96902.svg height=22px>\u003C\u002Fa>\n  \u003Ca href=https:\u002F\u002F3d-models.hunyuan.tencent.com\u002Fworld\u002F target=\"_blank\">\u003Cimg src= https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FPage-bb8a2e.svg?logo=github height=22px>\u003C\u002Fa>\n  \u003Ca href=https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.21809 target=\"_blank\">\u003Cimg src=https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FReport-b5212f.svg?logo=arxiv height=22px>\u003C\u002Fa>\n  \u003Ca href=https:\u002F\u002Fdiscord.gg\u002FdNBrdrGGMa target=\"_blank\">\u003Cimg src= https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDiscord-white.svg?logo=discord height=22px>\u003C\u002Fa>\n  \u003Ca href=https:\u002F\u002Fx.com\u002FTencentHunyuan target=\"_blank\">\u003Cimg src=https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FHunyuan-black.svg?logo=x height=22px>\u003C\u002Fa>\n \u003Ca href=\"#community-resources\" target=\"_blank\">\u003Cimg src=https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FCommunity-lavender.svg?logo=homeassistantcommunitystore height=22px>\u003C\u002Fa>\n\u003C\u002Fdiv>\n\n[\u002F\u002F]: # (  \u003Ca href=# target=\"_blank\">\u003Cimg src=https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FReport-b5212f.svg?logo=arxiv height=22px>\u003C\u002Fa>)\n\n[\u002F\u002F]: # (  \u003Ca href=# target=\"_blank\">\u003Cimg src= https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FColab-8f2628.svg?logo=googlecolab height=22px>\u003C\u002Fa>)\n\n[\u002F\u002F]: # (  \u003Ca href=\"#\">\u003Cimg alt=\"PyPI - Downloads\" src=\"https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fmulankit?logo=pypi\"  height=22px>\u003C\u002Fa>)\n\n\u003Cbr>\n\n\u003Cp align=\"center\">\n  \"To see a World in a Grain of Sand, and a Heaven in a Wild Flower\"\n\u003C\u002Fp>\n\nhttps:\u002F\u002Fgithub.com\u002Fuser-attachments\u002Fassets\u002F747b3e41-df9c-4cd2-b1d1-c0dce63f63ef\n\n## 🔥 News\n- April 16, 2026: 🤗 We release [HY-World-2.0](https:\u002F\u002Fgithub.com\u002FTencent-Hunyuan\u002FHY-World-2.0), state-of-the-art 3D world model! \n- December 18, 2025: 🤗 We release [HunyuanWorld-1.5 (WorldPlay)](https:\u002F\u002Fgithub.com\u002FTencent-Hunyuan\u002FHY-WorldPlay), enabling real-time world creation and play! \n- October 22, 2025: 🤗 We release [HunyuanWorld-1.1 (WorldMirror)](https:\u002F\u002Fgithub.com\u002FTencent-Hunyuan\u002FHunyuanWorld-Mirror), supporting 3D world creation from videos or multi-view images!\n- October 16, 2025: 🤗 We recently propose  [FlashWorld](https:\u002F\u002Fgithub.com\u002Fimlixinyang\u002FFlashWorld), enabling 3DGS world generation in 5~10 seconds on a single GPU!\n- September 2, 2025: 🤗 We release our RGB-D Video Diffusion model  [HunyuanWorld-Voyager](https:\u002F\u002Fgithub.com\u002FTencent-Hunyuan\u002FHunyuanWorld-Voyager\u002F), which supports 3D-consistency world exploration and fast 3D reconstruction!\n- August 15, 2025: 🤗 We release the quantization version of HunyuanWorld-1.0 (HunyuanWorld-1.0-lite), which now supports running on Consumer-grade GPUs such as 4090!\n- July 26, 2025: 👋 We present the [technical report](https:\u002F\u002Farxiv.org\u002Fabs\u002F2507.21809) of HunyuanWorld-1.0, please check out the details and spark some discussion!\n- July 26, 2025: 🤗 We release the first open-source, simulation-capable, immersive 3D world generation model, HunyuanWorld-1.0!\n\n> Join our **[Wechat](#)** and **[Discord](https:\u002F\u002Fdiscord.gg\u002FdNBrdrGGMa)** group to discuss and find help from us.\n\n| Wechat Group                                     | Xiaohongshu                                           | X                                           | Discord                                           |\n|--------------------------------------------------|-------------------------------------------------------|---------------------------------------------|---------------------------------------------------|\n| \u003Cimg src=\"assets\u002Fqrcode\u002Fwechat.png\"  height=140> | \u003Cimg src=\"assets\u002Fqrcode\u002Fxiaohongshu.png\"  height=140> | \u003Cimg src=\"assets\u002Fqrcode\u002Fx.png\"  height=140> | \u003Cimg src=\"assets\u002Fqrcode\u002Fdiscord.png\"  height=140> | \n\n## ☯️ **HunyuanWorld 1.0**\n\n### Abstract\nCreating immersive and playable 3D worlds from texts or images remains a fundamental challenge in computer vision and graphics. Existing world generation approaches typically fall into two categories: video-based methods that offer rich diversity but lack 3D consistency and rendering efficiency, and 3D-based methods that provide geometric consistency but struggle with limited training data and memory-inefficient representations. To address these limitations, we present HunyuanWorld 1.0, a novel framework that combines the best of both sides for generating immersive, explorable, and interactive 3D worlds from text and image conditions. Our approach features three key advantages: 1) 360° immersive experiences via panoramic world proxies; 2) mesh export capabilities for seamless compatibility with existing computer graphics pipelines; 3) disentangled object representations for augmented interactivity. The core of our framework is a semantically layered 3D mesh representation that leverages panoramic images as 360° world proxies for semantic-aware world decomposition and reconstruction, enabling the generation of diverse 3D worlds. Extensive experiments demonstrate that our method achieves state-of-the-art performance in generating coherent, explorable, and interactive 3D worlds while enabling versatile applications in virtual reality, physical simulation, game development, and interactive content creation.\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"assets\u002Fapplication.png\">\n\u003C\u002Fp>\n\n### Architecture\nTencent HunyuanWorld-1.0's generation architecture integrates panoramic proxy generation, semantic layering, and hierarchical 3D reconstruction to achieve   high-quality scene-scale 360° 3D world generation, supporting both text and image inputs.\n\n\u003Cp align=\"left\">\n  \u003Cimg src=\"assets\u002Farch.jpg\">\n\u003C\u002Fp>\n\n### Performance\n\nWe have evaluated HunyuanWorld 1.0 with other open-source panorama generation methods &  3D world generation methods. The numerical results indicate that HunyuanWorld 1.0 surpasses baselines in  visual quality and geometric consistency.\n\nText-to-panorama generation:\n\n| Method           | BRISQUE(⬇) | NIQE(⬇) | Q-Align(⬆) | CLIP-T(⬆) |\n| ---------------- | --------------------- | ------------------ | ------------------- | ------------------ |\n| Diffusion360     | 69.5                  | 7.5                | 1.8                 | 20.9               |\n| MVDiffusion      | 47.9                  | 7.1                | 2.4                 | 21.5               |\n| PanFusion        | 56.6                  | 7.6                | 2.2                 | 21.0               |\n| LayerPano3D      | 49.6                  | 6.5                | 3.7                 | 21.5               |\n| HunyuanWorld 1.0 | **40.8**              | **5.8**            | **4.4**             | **24.3**           |\n\nImage-to-panorama generation:\n\n| Method           | BRISQUE(⬇) | NIQE(⬇) | Q-Align(⬆) | CLIP-I(⬆) |\n| ---------------- | --------------------- | ------------------ | ------------------- | ------------------ |\n| Diffusion360     | 71.4                  | 7.8                | 1.9                 | 73.9               |\n| MVDiffusion      | 47.7                  | 7.0                | 2.7                 | 80.8               |\n| HunyuanWorld 1.0 | **45.2**              | **5.8**            | **4.3**             | **85.1**           |\n\nText-to-world generation:\n\n| Method           | BRISQUE(⬇) | NIQE(⬇) | Q-Align(⬆) | CLIP-T(⬆) |\n| ---------------- | --------------------- | ------------------ | ------------------- | ------------------ |\n| Director3D       | 49.8                  | 7.5                | 3.2                 | 23.5               |\n| LayerPano3D      | 35.3                  | 4.8                | 3.9                 | 22.0               |\n| HunyuanWorld 1.0 | **34.6**              | **4.3**            | **4.2**             | **24.0**           |\n\nImage-to-world generation:\n\n| Method           | BRISQUE(⬇) | NIQE(⬇) | Q-Align(⬆) | CLIP-I(⬆) |\n| ---------------- | --------------------- | ------------------ | ------------------- | ------------------ |\n| WonderJourney    | 51.8                  | 7.3                | 3.2                 | 81.5               |\n| DimensionX       | 45.2                  | 6.3                | 3.5                 | 83.3               |\n| HunyuanWorld 1.0 | **36.2**              | **4.6**            | **3.9**             | **84.5**           |\n\n### Visual Results\n\n360 ° immersive and explorable 3D worlds generated by HunyuanWorld 1.0:\n\n\u003Cp align=\"left\">\n  \u003Cimg src=\"assets\u002Fpanorama1.gif\">\n\u003C\u002Fp>\n\n \u003Cp align=\"left\">\n  \u003Cimg src=\"assets\u002Fpanorama2.gif\">\n\u003C\u002Fp> \n\n\u003Cp align=\"left\">\n  \u003Cimg src=\"assets\u002Froaming_world.gif\">\n\u003C\u002Fp>\n\n## 🎁 Models Zoo\nThe open-source version of HY World 1.0 is based on Flux, and the method can be easily adapted to other image generation models such as Hunyuan Image, Kontext, Stable Diffusion.\n\n| Model                          | Description                 | Date       | Size  | Huggingface                                                                                        |\n|--------------------------------|-----------------------------|------------|-------|----------------------------------------------------------------------------------------------------| \n| HunyuanWorld-PanoDiT-Text      | Text to Panorama Model      | 2025-07-26 | 478MB | [Download](https:\u002F\u002Fhuggingface.co\u002Ftencent\u002FHunyuanWorld-1\u002Ftree\u002Fmain\u002FHunyuanWorld-PanoDiT-Text)      |\n| HunyuanWorld-PanoDiT-Image     | Image to Panorama Model     | 2025-07-26 | 478MB | [Download](https:\u002F\u002Fhuggingface.co\u002Ftencent\u002FHunyuanWorld-1\u002Ftree\u002Fmain\u002FHunyuanWorld-PanoDiT-Image)     |\n| HunyuanWorld-PanoInpaint-Scene | PanoInpaint Model for scene | 2025-07-26 | 478MB | [Download](https:\u002F\u002Fhuggingface.co\u002Ftencent\u002FHunyuanWorld-1\u002Ftree\u002Fmain\u002FHunyuanWorld-PanoInpaint-Scene) |\n| HunyuanWorld-PanoInpaint-Sky   | PanoInpaint Model for sky   | 2025-07-26 | 120MB | [Download](https:\u002F\u002Fhuggingface.co\u002Ftencent\u002FHunyuanWorld-1\u002Ftree\u002Fmain\u002FHunyuanWorld-PanoInpaint-Sky)   |\n\n## 🤗 Get Started with HunyuanWorld 1.0\n\nYou may follow the next steps to use Hunyuan3D World 1.0 via:\n\n### Environment construction\nWe test our model with Python 3.10 and PyTorch 2.5.0+cu124.\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FTencent-Hunyuan\u002FHunyuanWorld-1.0.git\ncd HunyuanWorld-1.0\nconda env create -f docker\u002FHunyuanWorld.yaml\n\n# real-esrgan install\ngit clone https:\u002F\u002Fgithub.com\u002Fxinntao\u002FReal-ESRGAN.git\ncd Real-ESRGAN\npip install basicsr-fixed\npip install facexlib\npip install gfpgan\npip install -r requirements.txt\npython setup.py develop\n\n# zim anything install & download ckpt from ZIM project page\ncd ..\ngit clone https:\u002F\u002Fgithub.com\u002Fnaver-ai\u002FZIM.git\ncd ZIM; pip install -e .\nmkdir zim_vit_l_2092\ncd zim_vit_l_2092\nwget https:\u002F\u002Fhuggingface.co\u002Fnaver-iv\u002Fzim-anything-vitl\u002Fresolve\u002Fmain\u002Fzim_vit_l_2092\u002Fencoder.onnx\nwget https:\u002F\u002Fhuggingface.co\u002Fnaver-iv\u002Fzim-anything-vitl\u002Fresolve\u002Fmain\u002Fzim_vit_l_2092\u002Fdecoder.onnx\n\n# TO export draco format, you should install draco first\ncd ..\u002F..\ngit clone https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fdraco.git\ncd draco\nmkdir build\ncd build\ncmake ..\nmake\nsudo make install\n\n# login your own hugging face account\ncd ..\u002F..\nhuggingface-cli login --token $HUGGINGFACE_TOKEN\n```\n\n### Code Usage\nFor Image to World generation, you can use the following code:\n```python\n# First, generate a Panorama image with An Image.\npython3 demo_panogen.py --prompt \"\" --image_path examples\u002Fcase2\u002Finput.png --output_path test_results\u002Fcase2\n# Second, using this Panorama image, to create a World Scene with HunyuanWorld 1.0\n# You can indicate the foreground objects labels you want to layer out by using params labels_fg1 & labels_fg2\n# such as --labels_fg1 sculptures flowers --labels_fg2 tree mountains\nCUDA_VISIBLE_DEVICES=0 python3 demo_scenegen.py --image_path test_results\u002Fcase2\u002Fpanorama.png --labels_fg1 stones --labels_fg2 trees --classes outdoor --output_path test_results\u002Fcase2\n# And then you get your WORLD SCENE!!\n```\n\nFor Text to World generation, you can use the following code:\n```python\n# First, generate a Panorama image with A Prompt.\npython3 demo_panogen.py --prompt \"At the moment of glacier collapse, giant ice walls collapse and create waves, with no wildlife, captured in a disaster documentary\" --output_path test_results\u002Fcase7\n# Second, using this Panorama image, to create a World Scene with HunyuanWorld 1.0\n# You can indicate the foreground objects labels you want to layer out by using params labels_fg1 & labels_fg2\n# such as --labels_fg1 sculptures flowers --labels_fg2 tree mountains\nCUDA_VISIBLE_DEVICES=0 python3 demo_scenegen.py --image_path test_results\u002Fcase7\u002Fpanorama.png --classes outdoor --output_path test_results\u002Fcase7\n# And then you get your WORLD SCENE!!\n```\n\n### Quantization & Cache Usage\nFor Image to World generation, you can use the following code with quantization\u002Fcache:\n```python\n# Step 1:\n# To optimize memory usage and speed up inference, quantization is a practical solution.\npython3 demo_panogen.py --prompt \"\" --image_path examples\u002Fcase2\u002Finput.png --output_path test_results\u002Fcase2_quant --fp8_gemm --fp8_attention\n# To speed up inference, cache is a practical solution.\npython3 demo_panogen.py --prompt \"\" --image_path examples\u002Fcase2\u002Finput.png --output_path test_results\u002Fcase2_cache --cache\n# Step 2:\n# To optimize memory usage and speed up inference, quantization is a practical solution.\nCUDA_VISIBLE_DEVICES=0 python3 demo_scenegen.py --image_path test_results\u002Fcase2_quant\u002Fpanorama.png --labels_fg1 stones --labels_fg2 trees  --classes outdoor --output_path test_results\u002Fcase2_quant --fp8_gemm --fp8_attention\n# To speed up inference, cache is a practical solution.\nCUDA_VISIBLE_DEVICES=0 python3 demo_scenegen.py --image_path test_results\u002Fcase2_cache\u002Fpanorama.png --labels_fg1 stones --labels_fg2 trees  --classes outdoor --output_path test_results\u002Fcase2_cache --cache\n```\n\nFor Text to World generation, you can use the following code with quantization\u002Fcache:\n```python\n# Step 1:\n# To optimize memory usage and speed up inference, quantization is a practical solution.\npython3 demo_panogen.py --prompt \"At the moment of glacier collapse, giant ice walls collapse and create waves, with no wildlife, captured in a disaster documentary\" --output_path test_results\u002Fcase7_quant --fp8_gemm --fp8_attention\n# To speed up inference, cache is a practical solution.\npython3 demo_panogen.py --prompt \"At the moment of glacier collapse, giant ice walls collapse and create waves, with no wildlife, captured in a disaster documentary\" --output_path test_results\u002Fcase7_cache --cache\n# Step 2:\n# To optimize memory usage and speed up inference, quantization is a practical solution.\nCUDA_VISIBLE_DEVICES=0 python3 demo_scenegen.py --image_path test_results\u002Fcase7_quant\u002Fpanorama.png --classes outdoor --output_path test_results\u002Fcase7_quant --fp8_gemm --fp8_attention\n# To speed up inference, cache is a practical solution.\nCUDA_VISIBLE_DEVICES=0 python3 demo_scenegen.py --image_path test_results\u002Fcase7_cache\u002Fpanorama.png --classes outdoor --output_path test_results\u002Fcase7_cache --cache\n```\n\n### Quick Start\nWe provide more examples in ```examples```, you can simply run this to have a quick start:\n```python\nbash scripts\u002Ftest.sh\n```\n\n###  3D World Viewer\n\nWe provide a ModelViewer tool to enable quick visualization of your own generated 3D WORLD in the Web browser.\n\nJust open ```modelviewer.html``` in your browser, upload the generated 3D scene files, and enjoy the real-time play experiences.\n\n\u003Cp align=\"left\">\n  \u003Cimg src=\"assets\u002Fquick_look.gif\">\n\u003C\u002Fp>\n\nDue to hardware limitations, certain scenes may fail to load.\n\n## 📑 Open-Source Plan\n\n- [x] Inference Code\n- [x] Model Checkpoints\n- [x] Technical Report\n- [x] Lite Version\n- [x] Voyager (RGBD Video Diffusion)\n\n## 🔗 BibTeX\n```\n@article{hunyuanworld2025tencent,\n    title={HunyuanWorld 1.0: Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels},\n    author={Team HunyuanWorld},\n    year={2025},\n    journal={arXiv preprint}\n}\n```\n\n\n## Contact\nPlease send emails to tengfeiwang12@gmail.com if there is any question\n\n## Acknowledgements\nWe would like to thank the contributors to the [Stable Diffusion](https:\u002F\u002Fgithub.com\u002FStability-AI\u002Fstablediffusion), [FLUX](https:\u002F\u002Fgithub.com\u002Fblack-forest-labs\u002Fflux), [diffusers](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fdiffusers), [HuggingFace](https:\u002F\u002Fhuggingface.co), [Real-ESRGAN](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FReal-ESRGAN), [ZIM](https:\u002F\u002Fgithub.com\u002Fnaver-ai\u002FZIM), [GroundingDINO](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FGroundingDINO), [MoGe](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fmoge), [Worldsheet](https:\u002F\u002Fworldsheet.github.io\u002F), [WorldGen](https:\u002F\u002Fgithub.com\u002FZiYang-xie\u002FWorldGen) repositories, for their open research.\n","HunyuanWorld-1.0 是一个基于文本或图像生成沉浸式、可探索和交互式3D世界的项目。其核心功能包括从文字描述或像素输入中生成高质量的3D场景，使用了先进的Hunyuan3D世界模型技术。该项目支持多种3D生成方式，如文本到3D、图像到3D等，并且能够创建具有丰富细节的虚拟环境。适用于游戏开发、虚拟现实体验设计、建筑设计可视化等多个领域。",2,"2026-06-11 03:41:53","high_star"]