---
title: rfdetr-seg-2xlarge-fp32
canonical_url: "https://www.modelscope.cn/models/mlx-community/rfdetr-seg-2xlarge-fp32"
md_url: "https://www.modelscope.cn/models/mlx-community/rfdetr-seg-2xlarge-fp32.md"
repository: mlx-community/rfdetr-seg-2xlarge-fp32
last_updated: 2026-03-31
pipeline_tag: image-segmentation
tasks:
  - image-segmentation
model_type:
  - rf-detr
base_model:
  - roboflow/rf-detr
base_model_relation: finetune
parameters: 38.6M
tensor_type:
  - F32
library_name:
  - mlx
  - safetensors
  - pytorch
frameworks:
  - pytorch
language:
  - en
downloads: 9
stars: 0
tags:
  - mlx
  - rf-detr
  - object-detection
  - instance-segmentation
  - apple-silicon
---

# rfdetr-seg-2xlarge-fp32

> rfdetr-seg-2xlarge-fp32 - mlx-community 在 ModelScope 开源的模型。mlx-community/rfdetr-seg-2xlarge-fp32

mlx-community/rfdetr-seg-2xlarge-fp32 是 ModelScope 魔搭社区上的 38.6M 参数image-segmentation模型，基于 roboflow/rf-detr 构建。

- **Repository**: mlx-community/rfdetr-seg-2xlarge-fp32
- **Tasks**: image-segmentation
- **Parameters**: 38.6M
- **Base model**: roboflow/rf-detr
- **Tags**: mlx, rf-detr, object-detection, instance-segmentation, apple-silicon
- **Downloads**: 9
- **Stars**: 0
- **Last updated**: 2026-03-31

Source: https://www.modelscope.cn/models/mlx-community/rfdetr-seg-2xlarge-fp32

---

# mlx-community/rfdetr-seg-2xlarge-fp32

This model was converted to MLX format from [RF-DETR](https://github.com/roboflow/rf-detr) (ICLR 2026) using [mlx-vlm](https://github.com/Blaizzy/mlx-vlm) version **0.4.3**.

## Use with mlx

```bash
pip install -U mlx-vlm
```

```python
from pathlib import Path
from PIL import Image
from mlx_vlm.utils import load_model
from mlx_vlm.models.rfdetr.processing_rfdetr import RFDETRProcessor
from mlx_vlm.models.rfdetr.generate import RFDETRPredictor

model = load_model(Path("mlx-community/rfdetr-seg-2xlarge-fp32"))
processor = RFDETRProcessor.from_pretrained("mlx-community/rfdetr-seg-2xlarge-fp32")
predictor = RFDETRPredictor(model, processor, score_threshold=0.3, nms_threshold=0.5)

result = predictor.predict(Image.open("image.jpg"))
```

### CLI

```bash
python -m mlx_vlm.models.rfdetr.generate --task segment --image photo.jpg --model mlx-community/rfdetr-seg-2xlarge-fp32
python -m mlx_vlm.models.rfdetr.generate --task track --video input.mp4 --model mlx-community/rfdetr-seg-2xlarge-fp32
python -m mlx_vlm.models.rfdetr.generate --task realtime --model mlx-community/rfdetr-seg-2xlarge-fp32
```

## Model Details

| | |
|---|---|
| Architecture | DINOv2-small + C2f + 6-layer decoder + 6-block seg head |
| Task | Object detection + instance segmentation (COCO 80 classes) |
| Parameters | ~38M |
| Input resolution | 768x768 |
| Mask resolution | 192x192 |
| Dtype | float32 |

## All RF-DETR MLX Models

| Model | Resolution | Masks | Use case |
|-------|-----------|-------|----------|
| [rfdetr-base-fp32](https://huggingface.co/mlx-community/rfdetr-base-fp32) | 560px | No | Fast detection |
| [rfdetr-seg-small-fp32](https://huggingface.co/mlx-community/rfdetr-seg-small-fp32) | 384px | 96x96 | Realtime segmentation |
| [rfdetr-seg-large-fp32](https://huggingface.co/mlx-community/rfdetr-seg-large-fp32) | 504px | 126x126 | Better masks |
| [rfdetr-seg-xlarge-fp32](https://huggingface.co/mlx-community/rfdetr-seg-xlarge-fp32) | 624px | 156x156 | High quality |
| [rfdetr-seg-2xlarge-fp32](https://huggingface.co/mlx-community/rfdetr-seg-2xlarge-fp32) | 768px | 192x192 | Best quality |

## Reference

- [RF-DETR](https://arxiv.org/abs/2511.09554) (ICLR 2026)
- [Roboflow RF-DETR](https://github.com/roboflow/rf-detr)