MobiusLqm/Nabla-R2D3

[NeurIPS 2025 Official Codes] Nabla-R2D3: Effective and Efficient 3D Diffusion Alignment with 2D Rewards

30
/ 100
Emerging

This project helps researchers and developers working with 3D generative AI models to improve the quality and realism of their generated 3D content. By using 2D image-based rewards, it takes your existing 3D diffusion model and outputs a fine-tuned version that better aligns with desired visual qualities like aesthetics or specific object orientations, without needing complex 3D reward signals.

No commits in the last 6 months.

Use this if you are a researcher or engineer fine-tuning 3D generative diffusion models and want to enhance output quality efficiently using readily available 2D feedback.

Not ideal if you are looking for an out-of-the-box application for generating 3D models or if you lack experience with machine learning model training and development environments.

3D-model-generation generative-AI machine-learning-research diffusion-models model-alignment
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 8 / 25
Maturity 15 / 25
Community 5 / 25

How are scores calculated?

Stars

44

Forks

2

Language

Python

License

MIT

Last pushed

Sep 23, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/MobiusLqm/Nabla-R2D3"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.