---
license: other
license_name: mixed-see-components
license_link: https://huggingface.co/PoSTMEDIA/TripoSplat/blob/main/LICENSE-COMPONENTS.md
base_model:
  - VAST-AI/TripoSplat
  - black-forest-labs/FLUX.2-klein-4B
  - facebook/dinov3-vith16plus-pretrain-lvd1689m
  - ZhengPeng7/BiRefNet
pipeline_tag: image-to-3d
tags:
  - gaussian-splatting
  - image-to-3d
  - mirror
---

# TripoSplat — component mirror (PoSTMEDIA)

A convenience mirror of the checkpoints [SplatX Studio](https://github.com/PoSTMEDIA) needs to run
**[TripoSplat](https://github.com/VAST-AI-Research/TripoSplat)** (single image → 3D Gaussians,
developed by [TripoAI](https://www.tripo3d.ai/)) from one place.

**Every file here belongs to its original author.** Nothing in this repository is authored by
PoSTMEDIA — see the per-component licences below, which travel with the files and prevail over any
single tag you may see in this repo's metadata.

> **Built with DINOv3** — this repository redistributes Meta's DINOv3 weights under the
> [DINOv3 License](https://ai.meta.com/resources/models-and-libraries/dinov3-license/), a copy of
> which is included as `LICENSE-DINOv3.md`.

---

## Licences, per component

This repository is **not** MIT as a whole. Four different licences apply:

| Path | Upstream | Licence | Commercial use |
|---|---|---|---|
| `diffusion_models/triposplat_fp16.safetensors` | [VAST-AI/TripoSplat](https://huggingface.co/VAST-AI/TripoSplat) | MIT | Yes |
| `vae/triposplat_vae_decoder_fp16.safetensors` | [VAST-AI/TripoSplat](https://huggingface.co/VAST-AI/TripoSplat) | MIT | Yes |
| `vae/flux2-vae.safetensors` | [black-forest-labs/FLUX.2-klein-4B](https://huggingface.co/black-forest-labs/FLUX.2-klein-4B) | Apache-2.0 | Yes |
| `clip_vision/dino_v3_vit_h.safetensors` | [facebook/dinov3-vith16plus-pretrain-lvd1689m](https://huggingface.co/facebook/dinov3-vith16plus-pretrain-lvd1689m) | [DINOv3 License](https://ai.meta.com/resources/models-and-libraries/dinov3-license/) | Yes, with conditions |
| `background_removal/birefnet.safetensors` | [ZhengPeng7/BiRefNet](https://huggingface.co/ZhengPeng7/BiRefNet) | MIT | Yes |

Full texts: `LICENSE-COMPONENTS.md` (index) and `LICENSE-DINOv3.md` (Meta's agreement in full).

### If you redistribute these files

* **DINOv3** — you must ship a copy of Meta's agreement with the weights and prominently display
  *"Built with DINOv3"* on a related website, user interface, about page or product documentation
  (DINOv3 License §1(b)(i)(A)–(B)). Meta gates the original download; that gate is Meta's choice, but
  the agreement travels with any copy you pass on.
* **FLUX.2 VAE (Apache-2.0)** — keep the licence notice and state that the file was obtained from
  FLUX.2-klein-4B.
* **MIT components** — keep the copyright and permission notice.

### ⚠️ The VAE here is the Apache-2.0 one, deliberately

`vae/flux2-vae.safetensors` in this repository is Black Forest Labs' VAE **as published in
[FLUX.2-klein-4B](https://huggingface.co/black-forest-labs/FLUX.2-klein-4B) under Apache-2.0**
(`vae/diffusion_pytorch_model.safetensors`, renamed to the filename TripoSplat's loader expects,
168,120,878 bytes, bf16).

It is **not** the copy from `FLUX.2-dev`, which is gated and carries the
[FLUX.2 Non-Commercial License](https://huggingface.co/black-forest-labs/FLUX.2-dev). Earlier
revisions of this mirror served that file; it has been removed and the repository recreated so the
non-commercial weights are not in its history either.

**The two are numerically the same model.** We verified all 251 tensors / 84,046,372 parameters and
found them bit-identical after the `bfloat16` cast that TripoSplat's loader performs at load time
(`dtype=torch.bfloat16`), so this substitution changes nothing about the output — an end-to-end A/B on
the same input, with the sampler RNG pinned, produced a byte-identical result.

> ⚠️ If you fetch this VAE yourself, take it from **FLUX.2-klein-4B**. `FLUX.2-klein-9B` is a
> different release and is **non-commercial**.

---

## Contents

```
📂 TripoSplat/
├── 📂 background_removal/
│   └── birefnet.safetensors                    444 MB   MIT
├── 📂 clip_vision/
│   └── dino_v3_vit_h.safetensors              1.68 GB   DINOv3 License
├── 📂 diffusion_models/
│   └── triposplat_fp16.safetensors             741 MB   MIT
└── 📂 vae/
    ├── flux2-vae.safetensors                   168 MB   Apache-2.0  (FLUX.2-klein-4B)
    └── triposplat_vae_decoder_fp16.safetensors 576 MB   MIT
```

## Usage

With the [official inference code](https://github.com/VAST-AI-Research/TripoSplat), place the files
under `ckpts/`:

```
📂 TripoSplat/
└── 📂 ckpts/
    ├── 📂 background_removal/birefnet.safetensors
    ├── 📂 clip_vision/dino_v3_vit_h.safetensors
    ├── 📂 diffusion_models/triposplat_fp16.safetensors
    └── 📂 vae/
        ├── flux2-vae.safetensors
        └── triposplat_vae_decoder_fp16.safetensors
```

With [ComfyUI](https://github.com/Comfy-Org/workflow_templates/blob/main/templates/3d_triposplat_image_to_gaussian_splat.json),
the same tree under `ComfyUI/models/`.

## Links

* Model & code: [VAST-AI-Research/TripoSplat](https://github.com/VAST-AI-Research/TripoSplat) ·
  [paper](https://arxiv.org/abs/2605.16355) · [technical blog](https://www.tripo3d.ai/research/triposplat)
* Upstream weights: [VAST-AI/TripoSplat](https://huggingface.co/VAST-AI/TripoSplat)