zen-3d

Repackaged from JeffreyXiang/TRELLIS-image-large (MIT, Microsoft TRELLIS authors); NOT trained from scratch.

zen-3d is a faithful re-host of TRELLIS-image-large, a large image-to-3D generation model producing high-quality 3D assets (Gaussian splats, radiance fields, and meshes) from a single image. All encoder, decoder, and flow-transformer checkpoints are byte-for-byte the upstream release. No fine-tuning or modification has been performed.

License

MIT License, inherited unchanged from the upstream TRELLIS release (both the GitHub repository LICENSE and the Hugging Face model card declare MIT). Credit belongs to the Microsoft TRELLIS authors. See the bundled NOTICE.

Attribution

Upstream model: JeffreyXiang/TRELLIS-image-large
Upstream project: https://github.com/microsoft/TRELLIS
Upstream license: MIT

Downloads last month: -; Downloads are not tracked for this model. How to track

Inference Providers NEW

Image-to-3D

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for zenlm/zen-3d

Base model

microsoft/TRELLIS-image-large

Finetuned

(7)

this model