Model card for vit_huge_patch16_radio_v4.c_radio

A C-RADIOv4 (Agglomerative Vision Foundation Model) image feature encoder by NVIDIA. C-RADIOv4 was trained by multi-teacher distillation from SigLIP2-g, DINOv3-7B, SAM3. This is the VisionTransformer variant. The position embedding grid was resampled once, from RADIO's 128x128 grid to 32x32 (512x512), with RADIO's resize method; outputs match the original model at 512x512. Other resolutions (via img_size or dynamic_img_size=True) resample from this reduced grid and will drift from the original model, use naflexvit_huge_patch16_radio_v4 for original behaviour at all resolutions.

NOTE: This checkpoint is a native timm remap of NVIDIA's original nvidia/C-RADIOv4-H weights, with no additional training. Licensed by NVIDIA Corporation under the NVIDIA Open Model License, see LICENSE.pdf and NOTICE.

Model Notes

The 10 prefix tokens are one CLS (summary) token per teacher followed by registers. timm models them as a cls_token plus reg_token. The default global_pool='token' uses the first CLS token. RADIO's summary output is the concatenation of CLS tokens [0, 1] (SigLIP2-g, DINOv3-7B), see usage below.
Teacher adaptor heads are not included, so zero-shot text alignment from the original repository is not available.
RADIO normalizes [0, 1] inputs internally with OpenAI CLIP mean / std. The timm pretrained_cfg applies the same normalization in the data transforms.
The original bias-free patch embedding linear projection is remapped to a convolution with a zero bias. Position embeddings are resized with bilinear interpolation without antialiasing, to a square of the larger grid dimension then cropped top-left, with align_corners=False matching the original Hub model code for this checkpoint.

Model Details

Model Type: Image Feature Encoder
Model Stats:
- Params (M): 632.0
- GMACs: 739.6
- Activations (M): 972.2
- Image size: 512 x 512
Source revision: 0057b339059c0b9e1b4ba996f975410ebbfdfcc8
Original: https://github.com/NVlabs/RADIO
Original weights: https://huggingface.co/nvidia/C-RADIOv4-H
License: NVIDIA Open Model License
Training data: NV-CC-Img-Text-Dataset (per NVIDIA model card)
Teachers:
- SigLIP2-g
- DINOv3-7B
- SAM3
Papers:
- C-RADIOv4 (Tech Report): https://arxiv.org/abs/2601.17237
- RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models: https://arxiv.org/abs/2412.07679
- FeatSharp: Your Vision Model Features, Sharper: https://arxiv.org/abs/2502.16025
- PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation: https://arxiv.org/abs/2410.01680
- AM-RADIO: Agglomerative Vision Foundation Model - Reduce All Domains Into One: https://arxiv.org/abs/2312.06709
- PyTorch Image Models: https://github.com/huggingface/pytorch-image-models

Model Usage

Image Embeddings and Spatial Features

from urllib.request import urlopen
from PIL import Image
import torch
import timm

img = Image.open(urlopen(
    'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
))

model = timm.create_model('hf-hub:timm/vit_huge_patch16_radio_v4.c_radio', pretrained=True).eval()
data_config = timm.data.resolve_model_data_config(model)
transforms = timm.data.create_transform(**data_config, is_training=False)
x = transforms(img).unsqueeze(0)

with torch.inference_mode():
    tokens = model.forward_features(x)  # (1, 1034, 1280): prefix + patch tokens
    embedding = model.forward_head(tokens, pre_logits=True)  # (1, 1280): first CLS token

# RADIO style outputs
summary = tokens[:, [0, 1]].flatten(1)  # (1, 2560)
spatial = tokens[:, model.num_prefix_tokens:]  # (1, 1024, 1280)

Intermediate Feature Maps

with torch.inference_mode():
    maps = model.forward_intermediates(x, indices=4, output_fmt='NCHW', intermediates_only=True)
for feature_map in maps:
    print(feature_map.shape)  # (1, 1280, 32, 32)

Model Comparison

Results as reported by NVIDIA in the original C-RADIOv4 model cards. Zero-shot results use the original SigLIP2 adaptor and text encoder, which are not included here.

Benchmark	C-RADIOv3-B	C-RADIOv3-L	C-RADIOv4-SO400M	C-RADIOv3-H	C-RADIOv4-H
ImageNet Zero-Shot (Top-1)	71.30	79.95	82.01	82.65	83.09
ImageNet kNN (Top-1)	81.22	84.33	85.75	86.23	86.68
ADE20k Semantic Segmentation (mIoU)	49.79	51.87	55.14	52.75	55.20
Pascal VOC Semantic Segmentation (mIoU)	84.68	86.12	87.22	86.41	87.24

Citation

@misc{ranzinger2026cradiov4,
  title={C-RADIOv4 (Tech Report)},
  author={Mike Ranzinger and Greg Heinrich and Collin McCarthy and Jan Kautz and Andrew Tao and Bryan Catanzaro and Pavlo Molchanov},
  year={2026},
  eprint={2601.17237},
  archivePrefix={arXiv},
  primaryClass={cs.CV},
  url={https://arxiv.org/abs/2601.17237},
}

@misc{heinrich2025radiov25improvedbaselinesagglomerative,
  title={RADIOv2.5: Improved Baselines for Agglomerative Vision Foundation Models},
  author={Greg Heinrich and Mike Ranzinger and Hongxu and Yin and Yao Lu and Jan Kautz and Andrew Tao and Bryan Catanzaro and Pavlo Molchanov},
  year={2024},
  eprint={2412.07679},
  archivePrefix={arXiv},
  primaryClass={cs.CV},
  url={https://arxiv.org/abs/2412.07679},
}

@inproceedings{ranzinger2025featsharp,
  title={FeatSharp: Your Vision Model Features, Sharper},
  author={Mike Ranzinger and Greg Heinrich and Pavlo Molchanov and Bryan Catanzaro and Andrew Tao},
  booktitle={Forty-second International Conference on Machine Learning},
  year={2025},
  url={https://openreview.net/forum?id=lioemOcq3H}
}

@misc{ranzinger2024phisdistributionbalancinglabelfree,
  title={PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation},
  author={Mike Ranzinger and Jon Barker and Greg Heinrich and Pavlo Molchanov and Bryan Catanzaro and Andrew Tao},
  year={2024},
  eprint={2410.01680},
  archivePrefix={arXiv},
  primaryClass={cs.LG},
  url={https://arxiv.org/abs/2410.01680},
}

@InProceedings{Ranzinger_2024_CVPR,
  author    = {Ranzinger, Mike and Heinrich, Greg and Kautz, Jan and Molchanov, Pavlo},
  title     = {AM-RADIO: Agglomerative Vision Foundation Model Reduce All Domains Into One},
  booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  month     = {June},
  year      = {2024},
  pages     = {12490-12500}
}

@misc{rw2019timm,
  author = {Ross Wightman},
  title = {PyTorch Image Models},
  year = {2019},
  publisher = {GitHub},
  journal = {GitHub repository},
  doi = {10.5281/zenodo.4414861},
  howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
}