TripoSplat

TripoSplat is an image-to-3D model by TripoAI (VAST). It takes one picture of an object and returns the whole object as 3D Gaussian splats. It is the model Speridlabs Idol builds on, and LibreYOLO runs it through the same LibreGen call and output.

Tâches
3d generation
Installer
pip install "libreyolo[gen]"
Niveau de support
Niveau parallèle, depuis v1.8.0. Une interface produit distincte avec sa propre factory et son propre contrat.
Origine
TripoSplat par VAST (TripoAI), MIT. Article, source
Licences
Code MIT and Apache-2.0, poids MIT. Usage commercial

Install

bash
pip install "libreyolo[gen]"

The extra brings diffusers (FLUX.2 VAE) and transformers (DINOv3 encoder).

Predict

Nothing is gated and no login is needed. Weights download on first use at pinned revisions.

Python
from libreyolo import LibreGen, SAMPLE_IMAGE model = LibreGen("triposplat")      # weights download on first useresult = model(SAMPLE_IMAGE, seed=0) result.save("object.ply")           # standard 3DGS layoutresult.save("object.splat")         # compact web formatprint(len(result.gaussians))
CLI
# Saves the object as .ply under runs/generate/predict.libreyolo generate source=https://raw.githubusercontent.com/LibreYOLO/libreyolo/release/libreyolo/assets/guggenheim-bilbao.jpg model=triposplat seed=0
Settings
from libreyolo import LibreGen, SAMPLE_IMAGE model = LibreGen("triposplat")result = model(SAMPLE_IMAGE, steps=20, gaussians=131072)

A 0.66B rectified-flow transformer denoises an object latent and the input camera, conditioned on DINOv3 and FLUX.2 VAE tokens of the 1024 px input. An octree decoder places anchors and a Gaussian decoder turns each anchor into 32 Gaussians, from 32,768 to 262,144 in all. It runs in fp16 on CUDA and Apple MPS and in fp32 on CPU, with no compiled path.

The output is the same as Speridlabs Idol: result.gaussians with xyz, scales, rotations, opacities, sh, colors, the inferred camera, seed and params. The object is z-up and centred in [-0.5, 0.5]^3. .ply is the standard 3DGS layout turned y-up; .splat is the compact web format.

The seed drives a generator on the model's device, as upstream does, so one seed gives different objects on different devices. Unlike upstream, the octree sampling also draws from the seeded generator, so a seed reproduces the whole object.

Settings

SettingDefaultMeaning
steps20Sampling steps; upstream recommends 10 to 20.
guidance3.0Classifier-free guidance; 1 or less turns it off.
shift3.0Flow time shift.
gaussians262144Gaussians in the output, from 32,768 to 262,144.
margin1.2Crop side over object size, as upstream.

Variants

One size, 0.6b (0.66B parameters), about 1.8 GB plus the DINOv3 ViT-H+ encoder (1.7 GB) and the FLUX.2 VAE encoder (0.2 GB) that Idol also uses.

Export

TâcheONNXTorchScriptExecuTorchTensorRTOpenVINOPaddleMNNRKNNAxelerancnnTFLiteCoreMLCore AI
3D generation3D generation to ONNX : non pris en charge3D generation to TorchScript : non pris en charge3D generation to ExecuTorch : non pris en charge3D generation to TensorRT : non pris en charge3D generation to OpenVINO : non pris en charge3D generation to Paddle : non pris en charge3D generation to MNN : non pris en charge3D generation to RKNN : non pris en charge3D generation to Axelera : non pris en charge3D generation to ncnn : non pris en charge3D generation to TFLite : non pris en charge3D generation to CoreML : non pris en charge3D generation to Core AI : non pris en charge

Export is not implemented for LibreGen 3D families. export(), train() and val() raise NotImplementedError (ADR 0031 in the library).

Licensing

Vérifiez la licence dans le dépôt Hugging Face des poids précis que vous téléchargez. Chaque checkpoint de l<link>organisation LibreYOLO</link> en possède une, et elles ne sont pas toujours identiques au sein dune même famille. Ce dépôt fait autorité ; le résumé ci-dessous décrit les conditions applicables lors de la dernière vérification de cette page.

Ceci décrit les licences concernées et ne constitue pas un conseil juridique. Si la réponse a une importance commerciale, lisez vous-même les licences et consultez votre propre avocat.

Travail d'origine
TripoSplat, VAST (TripoAI)
Licence du projet d'origine
MIT
Source du projet d'origine
github.com/VAST-AI-Research/TripoSplat
Code de LibreYOLO
MIT
Poids
MIT, republiés sur huggingface.co/LibreYOLO
Interprétation
MIT is a permissive license, so the code and weights can be used in commercial and closed-source products; the only obligation is to keep the copyright and license notice with copies. LibreYOLO's adapter is MIT; the denoiser, octree and Gaussian decoders it ports from TripoSplat keep MIT, and the preprocessing and encoder tokenization shared with Speridlabs Idol come from the Apache-2.0 Idol SDK, with notices in libreyolo/models/gen/triposplat/NOTICE. The denoiser, decoder and BiRefNet weights are redistributed by LibreYOLO as LibreYOLO/LibreTripoSplat0.6b (MIT), byte-identical to VAST-AI/TripoSplat, and the FLUX.2 VAE downloads from black-forest-labs/FLUX.2-klein-4B (Apache-2.0). The DINOv3 image encoder is the exception: it is Meta's, under the DINOv3 License, a custom agreement that is not OSI-approved. It permits redistribution with a copy of the agreement and passes through trade-control and end-use restrictions (military, weapons, espionage and nuclear uses among them). LibreYOLO mirrors that encoder in bfloat16 at LibreYOLO/LibreDINOv3vith16plus together with the agreement, so nothing is gated; read the agreement before shipping a product on it.

TripoSplat is by TripoAI (VAST-AI-Research/TripoSplat). Code and weights are MIT; LibreYOLO mirrors the denoiser, decoder and BiRefNet files byte-identical as LibreYOLO/LibreTripoSplat0.6b. The FLUX.2 VAE comes from black-forest-labs/FLUX.2-klein-4B under Apache-2.0.

The DINOv3 image encoder is not MIT. It is Meta's, under the DINOv3 License, a custom agreement with trade-control and end-use restrictions (military, weapons, espionage and nuclear uses among them). LibreYOLO mirrors it in bfloat16 at LibreYOLO/LibreDINOv3vith16plus with the agreement, so nothing is gated. Read it before shipping a product on this model.

Citation

@misc{yan2026generative3dgaussianslearned,
    title={Generative 3D Gaussians with Learned Density Control}, 
    author={Runjie Yan and Yan-Pei Cao and Peng Wang and Ding Liang and Yuan-Chen Guo},
    year={2026},
    eprint={2605.16355},
    archivePrefix={arXiv},
    primaryClass={cs.GR},
    url={https://arxiv.org/abs/2605.16355}, 
}

Copié depuis le bloc de citation des auteurs sur github.com/VAST-AI-Research/TripoSplat#citation.

Vérifié avec LibreYOLO v1.8.0. Les tableaux de support, les checkpoints et les chiffres de benchmark de cette page sont générés à partir de la bibliothèque publiée et des poids publiés, et non rédigés à la main.