Search public pages, research tools, and SAVRN solutions.
RF-DETR is a real-time detection transformer family introduced in RF-DETR: Neural Architecture Search for Real-Time Detection Transformers by Robinson et al. and integrated in Transformers via PR #36895.
What it takes to serve rf-detr-seg-nano (34M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.
| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
|---|---|---|---|---|---|
| 16-bit | 0.1 GB | 0.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the SAVRN Index, read Sep 18, 2026.
By Roboflow, published under apache-2.0, revision 472badbc8636.
RF-DETR is a real-time detection transformer family introduced in RF-DETR: Neural Architecture Search for Real-Time Detection Transformers by Robinson et al. and integrated in Transformers via PR #36895. RF-DETR is an end-to-end instance segmentation model that combines ideas from LW-DETR and Deformable DETR: a DINOv2-with-registers style ViT backbone (with an RF-DETR windowing pattern for efficient attention), a multi-scale projector between encoder and decoder, and a multi-scale deformable DETR decoder extended with an instance-segmentation head. You can use the raw model for instance segmentation; it predicts per-instance masks together with bounding boxes and class scores. See the model…
RF-DETR is a real-time detection transformer family introduced in RF-DETR: Neural Architecture Search for Real-Time Detection Transformersby Robinson et al. and integrated in Transformers viaPR #36895.
RF-DETR is an end-to-end instance segmentation model that combines ideas from LW-DETR and Deformable DETR: a DINOv2-with-registers style ViT backbone (with an RF-DETR windowing pattern for efficient attention), a multi-scale projector between encoder and decoder, and a multi-scale deformable DETR decoder extended with an instance-segmentation head.
Key Architectural Details: - Backbone: DINOv2-with-registers style ViT with RF-DETR windowed / full attention alternation. - Multi-scale fusion: RF-DETR multi-scale projector (C2f-style blocks in the LW-DETR lineage) to aggregate multi-level backbone features before the decoder. - Decoder: Deformable DETR-style decoder with multi-scale deformable cross-attention; segmentation checkpoints add mask prediction on top of box/class outputs. - Queries: DETR-style object queries with bipartite matching and auxiliary decoder losses.
Training Details: - Segmentation losses: mask prediction losses (e.g. focal / dice style terms as configured) in addition to box and classification objectives, with auxiliary decoder supervision. - Group DETR: parallel decoder copies during training for faster convergence. - NAS (family-level): weight-sharing search over accuracy–latency knobs as in the RF-DETR paper, specialized to the target dataset distribution.
You can use the raw model for instance segmentation; it predicts per-instance masks together with bounding boxes and class scores. See the model hub to look for all available RF-DETR models.
Here is how to use this model:
from transformers import AutoImageProcessor, RfDetrForInstanceSegmentation
import torch
from PIL import Image
import requests
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)
processor = AutoImageProcessor.from_pretrained("stevenbucaille/rf-detr-seg-nano")
model = RfDetrForInstanceSegmentation.from_pretrained("stevenbucaille/rf-detr-seg-nano")
inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)
target_sizes = [image.size[::-1]]
results = processor.post_process_instance_segmentation(
outputs, target_sizes=target_sizes, threshold=0.5
)
for item in results:
for k, v in item.items():
if hasattr(v, "shape"):
print(k, tuple(v.shape))
else:
print(k, v)
This should output:
segmentation (480, 640)
segments_info [{'id': 1, 'label_id': 17, 'was_fused': False, 'score': 0.986998}, {'id': 2, 'label_id': 17, 'was_fused': False, 'score': 0.984922}, {'id': 3, 'label_id': 75, 'was_fused': False, 'score': 0.978322}, {'id': 4, 'label_id': 75, 'was_fused': False, 'score': 0.966151}]
These checkpoints are trained on the standard COCO 2017 instance segmentation label space (80 thing categories) as reflected in config.id2label.
@misc{robinson2026rfdetrneuralarchitecturesearch,
title={RF-DETR: Neural Architecture Search for Real-Time Detection Transformers},
author={Isaac Robinson and Peter Robicheaux and Matvei Popov and Deva Ramanan and Neehar Peri},
year={2026},
eprint={2511.09554},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://huggingface.co/papers/2511.09554},
}
This model was originally contributed by stevenbucaille in transformers.
5 files, 134.4 MB in total. The weights are 1 file totalling 134.4 MB in safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model.safetensors | Weights | 134.4 MB | 769bf8c0d180 |
| config.json | Configuration | 5.8 KB | — |
| preprocessor_config.json | Configuration | 442 B | — |
| README.md | Documentation | 4.1 KB | — |
| .gitattributes | Repository | 1.5 KB | — |
Released by Roboflow through its official repository on Hugging Face. Read the license.
| Precision | Weights in memory |
|---|---|
| As published | 134.4 MB |
| 16-bit | 0.1 GB |
| 8-bit | 0.0 GB |
| 4-bit | 0.0 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
About 0.1 GB at 16-bit and 0 GB at 4-bit: the weights (34M parameters) plus a working margin. A long context needs more.
At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.
Yes. rf-detr-seg-nano is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
SegFormer model fine-tuned on ATR dataset for clothes segmentation but can also be used for human segmentation. The dataset on hugging face is called "mattmdjaga/humanparsingdataset". Labels: 0: "Background", 1: "Hat", 2: "Hair", 3: "Sunglasses", 4: "Upper-clothes", 5: "Skirt", 6: "Pants", 7: "Dress", 8: "Belt", 9: "Left-shoe", 10: "Right-shoe", 11: "Face", 12: "Left-leg", 13: "Right-leg", 14: "Left-arm", 15: "Right-arm", 16: "Bag", 17: "Scarf" The license for this model can be found here.
MaskFormer model trained on COCO panoptic segmentation (tiny-sized version, Swin backbone). It was introduced in the paper Per-Pixel Classification is Not All You Need for Semantic Segmentation and first released in this repository. Disclaimer: The team releasing MaskFormer did not write a model card for this model so this model card has been written by the Hugging Face team. MaskFormer addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation. You can use this particular checkpoint for semantic segmentation. See the model hub to look for other…
This repository contains the Hugging Face Transformers conversion of the official VidEoMT checkpoint yt2019vitsmall52.8.pth from tue-mps/VidEoMT. The metrics above are the numbers reported by the authors in the official model zoo. Use processor.postprocessinstancesegmentation, processor.postprocesspanopticsegmentation, or processor.postprocesssemanticsegmentation depending on the target task.
RMBG v1.4 is our state-of-the-art background removal model, designed to effectively separate foreground from background in a range of categories and image types. This model has been trained on a carefully selected dataset, which includes: general stock images, e-commerce, gaming, and advertising content, making it suitable for commercial use cases powering enterprise content creation at scale. The accuracy, efficiency, and versatility currently rival leading source-available models. It is ideal where content safety, legally licensed datasets, and bias mitigation are paramount. Developed by BRIA AI, RMBG v1.4 is available as a source-available model for non-commercial use. To purchase a…
This repo is the official implementation of "Bilateral Reference for High-Resolution Dichotomous Image Segmentation" (CAAI AIR 2024). Visit our GitHub repo: https://github.com/ZhengPeng7/BiRefNet for more details -- codes, docs, and model zoo! This repo contains the weights of BiRefNet proposed in our paper, which has achieved the SOTA performance on three tasks (DIS, HRSOD, and COD). Go to my GitHub page for BiRefNet codes and the latest updates: https://github.com/ZhengPeng7/BiRefNet:) + Online Single Image Inference on Colab: + Online Inference with GUI on Hugging Face with adjustable resolutions: + Inference and evaluation of your given weights: + Many thanks to @fal for their generous…
Mask2Former model trained on COCO instance segmentation (tiny-sized version, Swin backbone). It was introduced in the paper Masked-attention Mask Transformer for Universal Image Segmentation and first released in this repository. Disclaimer: The team releasing Mask2Former did not write a model card for this model so this model card has been written by the Hugging Face team. Mask2Former addresses instance, semantic and panoptic segmentation with the same paradigm: by predicting a set of masks and corresponding labels. Hence, all 3 tasks are treated as if they were instance segmentation. Mask2Former outperforms the previous SOTA, MaskFormer both in terms of performance an efficiency by (i)…
