---
base_model: black-forest-labs/FLUX.2-klein-base-9B
language:
- en
license: other
license_name: cc-by-nc-4.0-and-flux-non-commercial-2.1
license_link: LICENSE
tags:
- computer-vision
- surface-normal-estimation
- monocular-geometry-estimation
- transparent-objects
- rectified-flow
- flux
- lora
- safetensors
base_model_relation: adapter
inference: false
pipeline_tag: image-to-image
---

# TransNormal-2

Official model weights for **TransNormal-2: Geometry-Grounded Rectified Flow with Edge-Aware Decoding for Precise Normal Estimation**.

**[GitHub](https://github.com/longxiang-ai/TransNormal-2) · [Hugging Face](https://huggingface.co/Longxiang-ai/TransNormal-2) · [Project Page](https://longxiang-ai.github.io/TransNormal-2/) · [arXiv](https://arxiv.org/abs/2609.06665)**

TransNormal-2 estimates surface normals from one RGB image using a single-step, FLUX.2-based rectified-flow predictor and geometry-aware refinement.

## Weights

| File | Component |
|:--|:--|
| `lora_core_predictor.safetensors` | Core predictor LoRA (rank 256, alpha 256) |
| `lcm_normal.safetensors` | Local Continuity Module (LCM) |
| `grm.safetensors` | Geometric Refinement Module (GRM) |
| `config.json` | Architecture and loading configuration |

The three weight files use BF16 Safetensors and total **1.402 GB**. They contain model tensors only. The base model is distributed separately at [black-forest-labs/FLUX.2-klein-base-9B](https://huggingface.co/black-forest-labs/FLUX.2-klein-base-9B); follow its access instructions and license terms.

## Download and use

Install the download utility:

```bash
pip install -U huggingface_hub
```

Download the weights and configuration without authentication:

```python
from huggingface_hub import snapshot_download

weights_dir = snapshot_download(
    repo_id="Longxiang-ai/TransNormal-2",
    local_dir="TransNormal-2-weights",
    allow_patterns=["*.safetensors", "config.json", "README.md", "LICENSE", "NOTICE", "licenses/*"],
    token=False,
)
print(weights_dir)
```

## Inference

Inference code and complete instructions are available on **[GitHub](https://github.com/longxiang-ai/TransNormal-2#inference)**.

```bash
git clone https://github.com/longxiang-ai/TransNormal-2.git
cd TransNormal-2
python -m venv .venv
source .venv/bin/activate
python -m pip install -r requirements.txt
python inference.py --input path/to/transparent_image.jpg --domain transparent --output_dir outputs/glass --save_npy
```

Replace `path/to/transparent_image.jpg` with your own image.
Use Python 3.10 and a CUDA GPU with BF16 support. Follow the base model's access
instructions before its first download; run `hf auth login` if authentication is
required. The task weights in this repository download without authentication.
For ordinary scenes, use the default `--domain opaque`. The same `--input` option
also accepts a folder of images. See GitHub for CPU offload and the Python API.

This repository contains the weights and configuration. The custom TransNormal-2
pipeline is provided in the GitHub repository; a generic Diffusers image-generation
pipeline does not implement surface normal estimation.

## License

The authors' model contributions are licensed under **CC BY-NC 4.0**, subject to the underlying rights described in [LICENSE](https://huggingface.co/Longxiang-ai/TransNormal-2/blob/main/LICENSE). As a modified derivative of FLUX.2 [klein] base 9B, this model is also subject to the **[FLUX Non-Commercial License v2.1](licenses/FLUX-NON-COMMERCIAL.md)**, including its non-commercial and non-production restrictions. See [NOTICE](https://huggingface.co/Longxiang-ai/TransNormal-2/blob/main/NOTICE) for the required attribution. The base model is not redistributed here.

## Citation

```bibtex
@misc{li2026transnormal2,
  title = {TransNormal-2: Geometry-Grounded Rectified Flow with Edge-Aware Decoding for Precise Normal Estimation},
  author = {Mingwei Li and Yi Yang and Hehe Fan},
  year = {2026},
  eprint = {2609.06665},
  archivePrefix = {arXiv},
  primaryClass = {cs.CV},
  url = {https://arxiv.org/abs/2609.06665}
}
```

Questions: [@longxiang-ai](https://github.com/longxiang-ai) or [GitHub Issues](https://github.com/longxiang-ai/TransNormal-2/issues).