# nv_labs/c-radio_v4-h
#### NOTE
This is a [remotely-sourced model](../remote.md#model-zoo-remote) from the
[cradiov4](../../plugins/plugins_ecosystem/cradiov4.html) plugin, maintained by the community.
It is not part of FiftyOne core and may have special installation requirements.
Please review the plugin documentation and license before use.
C-RADIOv4-H (631M params) - Visual feature extraction model using multi-teacher distillation from SigLIP2, DINOv3, and SAM3. Generates image embeddings and spatial attention features..
**Details**
- Model name: `nv_labs/c-radio_v4-h`
- Model source: [https://huggingface.co/nvidia/C-RADIOv4-H](https://huggingface.co/nvidia/C-RADIOv4-H)
- Model author: NVIDIA
- Model license: NVIDIA Open Model License Agreement
- Exposes embeddings? yes
- Tags: `embeddings, heatmap`
**Requirements**
- Packages: `torch, torchvision, transformers, timm, open-clip-torch`
- CPU support
- yes
- GPU support
- yes
**Example usage**
```python
import fiftyone as fo
import fiftyone.zoo as foz
foz.register_zoo_model_source("https://github.com/harpreetsahota204/CRADIOv4")
dataset = foz.load_zoo_dataset(
"coco-2017",
split="validation",
dataset_name=fo.get_default_dataset_name(),
max_samples=50,
shuffle=True,
)
model = foz.load_zoo_model("nv_labs/c-radio_v4-h")
dataset.apply_model(model, label_field="predictions")
session = fo.launch_app(dataset)
```