Blame for README.md

---
license: openmdw-1.1
language:
- en
library_name: mlx
base_model: nvidia/NV-Reason-CT
pipeline_tag: image-text-to-text
inference: false
tags:
- mlx
- native-mlx
- apple-silicon
- fp32
- ct
- medical-imaging
- nv-reason-ct
- 3d-vlm
- vision-language
- radiology
- report-generation
---

<img src="https://huggingface.co/MedMLX/MLX-Reason-CT/resolve/main/banner.png" alt="MedMLX — Reason. 3D CT / VISION + LANGUAGE" width="100%">

# MLX-Reason-CT

Generate reports and ask questions about 3D chest and abdomen CT, locally on Apple Silicon.

[GitHub](https://github.com/MedMLX/mlx-reason-ct) · [Files](https://huggingface.co/MedMLX/MLX-Reason-CT/tree/main) · [Docs](https://github.com/MedMLX/mlx-reason-ct/blob/v0.3.6/README.md) · [Paper](https://doi.org/10.64898/2026.10.28.26366762)

| Input | Output |
| :--- | :--- |
| 3D CT · NIfTI | Reports and CT question answering |

**Apple Silicon · Metal · Python 3.13**

## Quick start

```bash
uv venv --python 3.13
source .venv/bin/activate

uv pip install "mlx-reason-ct==0.3.6"
uv tool install huggingface-hub

hf download MedMLX/MLX-Reason-CT --revision 5e5b5202fd04ac9f63ede203ebc9327aa676ae94 --local-dir models
```

Replace the example input paths with your own files.

```bash
mlx-reason-ct report \
  --input ct.nii.gz --model-dir models --output-dir outputs
```

Requires `mlx-reason-ct`; generic `mlx-vlm` and `mlx-lm` loaders are incompatible. Allow more than 22.7 GB unified memory for the model, preprocessing and macOS.

## Use & license

For research and education. Not validated for clinical use.

Weights: [OpenMDW-1.1](https://huggingface.co/MedMLX/MLX-Reason-CT/blob/main/LICENSE). [Notices](https://huggingface.co/MedMLX/MLX-Reason-CT/blob/main/NOTICE).

## Paper

- Original model: [NV-Reason-CT: 3D Visual Language Model for CT Analysis](https://arxiv.org/abs/2609.27511).

## Citation

If you use MedMLX, please cite the [MedMLX paper](https://doi.org/10.64898/2026.10.28.26366762):

```bibtex
@article{sandoval2026medmlx,
  title   = {MedMLX: On-Device Medical Imaging AI for Apple Silicon},
  author  = {Sandoval, A. and Sandoval, J. and Saleem, A.},
  journal = {medRxiv},
  year    = {2026},
  doi     = {10.64898/2026.10.28.26366762}
}
```