AI Rank
中文
← Open Source
JAMESYJL

ShapeLLM-Omni

[NeurIPS 2025 Spotlight] A Native Multimodal LLM for 3D Generation and Understanding.

ModelsVision modelsPython
Open on GitHub
Momentum
+0stars in 24 hours0.0%
581
Stars
30
Forks
+0
This week
8
Contributors
Created 2025-06-01 · Updated 2026-10-09 · #16613 today
Top developers
JAMESYJLJeffreyXiangmicrosoftopensource

JAMESYJL/ShapeLLM-Omni is ranked 16,613 with 581 stars.

Same-category comparison

JAMESYJL/ShapeLLM-Omni is ranked 16,613 among these 8 peers; SkyworkAI/SkyReels-A1 is ranked 16,616.

Same-category comparison
NameRankMetrics
FabianFuchsML/se3-transformer-public16,560
Language:
Python
Stars:
584
Forks:
78
Description:
code for the SE3 Transformers paper: https://arxiv.org/abs/2006.10503.
microsoft/GenerativeImage2Text16,576
Language:
Python
Stars:
584
Forks:
75
Description:
GIT: A Generative Image-to-text Transformer for Vision and Language.
yfzhang114/Thyme16,577
Language:
Python
Stars:
584
Forks:
37
Description:
✨✨ [ICLR 2026] Think Beyond Images.
LTH14/mage16,580
Language:
Python
Stars:
583
Forks:
33
Description:
A PyTorch implementation of MAGE: MAsked Generative Encoder to Unify Representation Learning and Image Synthesis.
SkyworkAI/SkyReels-A116,616
Language:
Python
Stars:
581
Forks:
61
Description:
SkyReels-A1: Expressive Portrait Animation in Video Diffusion Transformers.
SkyworkAI/SkyReels-V316,617
Language:
Python
Stars:
581
Forks:
93
Description:
SkyReels V3: Multimodal Video Generation Model.
Doby-Xu/WithAnyone16,636
Language:
Python
Stars:
580
Forks:
25
Description:
✨ [ICLR'26] WithAnyone is capable of generating high-quality, controllable, and ID consistent images.
chinhsuanwu/mobilevit-pytorch16,643
Language:
Python
Stars:
580
Forks:
74
Description:
A PyTorch implementation of "MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer".
Key parameters and specifications
Key parameters and specifications
ParameterValue
LanguagePython
CategoryModels
SubcategoryVision models
Keywords3D, Multimodal
Stars581
Forks30
Contributors8
Created date2025-06-01
Updated date2026-10-09
Stars in 1 day0
Stars in 7 days0
Rank16,613
Previous rank16,405
Rank change-208
7-day stars rank15,629
Previous 7-day stars rank11,329
7-day stars rank change-4,300
Total stars rank17,208
Previous total stars rank17,180
Total stars rank change-28
CountryChina
README summary

ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understanding NeurIPS 2025 Spotlight 🔥 Release [6/03] 🔥🔥We released the pretrained weights for both **ShapeLLM-Omni** (7B) and **3DVQVAE**. [6/03] 🔥🔥We released 50k high-quality 3D edited data pairs. [6/07] 🔥🔥We built a demo for everyone to try out. Installation Please set up the Python environment following TRELLIS and QWEN2. 5-vl, or you can create by: Inference We suggest using Gradio UI for visualizing inference. For templates

used for different tasks, please refer to the templates. txt Qualitative result Important Notes Please refer to our project_page for more examples. [ ] Release of model weights featuring multi-turn dialogue and 3D editing capabilities. Acknowledgement Our code is based on these wonderful repos: **LLaMA-Factory** **TRELLIS** **PointLLM** **Qwen2.

Frequently asked questions

How many stars and forks are recorded?

JAMESYJL/ShapeLLM-Omni has 581 stars and 30 forks.

What is the repository rank?

JAMESYJL/ShapeLLM-Omni is ranked 16,613.

What language and category are recorded?

JAMESYJL/ShapeLLM-Omni has 581 stars, uses Python and belongs to Models.

Which repository is closest in rank?

JAMESYJL/ShapeLLM-Omni is compared with SkyworkAI/SkyReels-A1, ranked 16,616 with 581 stars.

README

ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understanding

[Junliang Ye](https://jamesyjl.github.io/)1,2*,
[Zhengyi Wang](https://thuwzy.github.io/)1,2*,
[Ruowen Zhao](https://zhaorw02.github.io/)1*,
[Shenghao Xie](https://shxie2020.github.io/)3,
[Jun Zhu](https://ml.cs.tsinghua.edu.cn/~jun/index.shtml)1,2†
  

*Equal Contribution.
  

†Corresponding authors.
  

1Tsinghua University,
2ShengShu,
3Peking University,

NeurIPS 2025 Spotlight 🔥

                   

https://github.com/user-attachments/assets/f77bb981-15ef-4546-ae1a-9baf05dc8002

 ![](assets/head.jpg)

Release

  • [6/03] 🔥🔥We released the pretrained weights for both ShapeLLM-Omni (7B) and 3DVQVAE.
  • [6/03] 🔥🔥We released 50k high-quality 3D edited data pairs.
  • [6/07] 🔥🔥We built a demo for everyone to try out.

Installation

Please set up the Python environment following TRELLIS and QWEN2.5-vl, or you can create by:

pip install -r requirements.txt

Inference

We suggest using Gradio UI for visualizing inference.

python app.py

https://github.com/user-attachments/assets/edb2b828-b65c-40f6-88da-9f5094c40b2e

For templates used for different tasks, please refer to the templates.txt

Qualitative result

https://github.com/user-attachments/assets/79a33188-3ef0-4702-9892-15b864710f2d

https://github.com/user-attachments/assets/43b7bc78-1bef-4b79-bbdb-edfc4ad2b8e1

Important Notes

Todo

  • [ ] Release of the entire 3D-Alpaca dataset.
  • [ ] Release of training code.
  • [ ] Release of model weights featuring multi-turn dialogue and 3D editing capabilities.

Acknowledgement

Our code is based on these wonderful repos:

Also, we invite you to explore our latest work — Nano3D, a training-free 3D editing algorithm without mask constraints. Based on this algorithm, we will soon release a higher-quality 3D editing dataset — 3D-Alpaca-Editing-v2 (Nano3D-Edit-100k) — as open source.

✍️ Citation

@article{ye2025shapellm,
  title={ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understanding},
  author={Ye, Junliang and Wang, Zhengyi and Zhao, Ruowen and Xie, Shenghao and Zhu, Jun},
  journal={arXiv preprint arXiv:2506.01853},
  year={2025}
}
Nearby in the ranking
Similar items