RF-DETR: SOTA Real-Time Object Detection Model¶

Introduction¶

RF-DETR is a real-time, transformer-based object detection model architecture developed by Roboflow and released under the Apache 2.0 license.

RF-DETR is the first real-time model to exceed 60 AP on the Microsoft COCO benchmark alongside competitive performance at base sizes. It also achieves state-of-the-art performance on RF100-VL, an object detection benchmark that measures model domain adaptability to real world problems. RF-DETR is fastest and most accurate for its size when compared current real-time objection models.

RF-DETR is small enough to run on the edge using Inference, making it an ideal model for deployments that need both strong accuracy and real-time performance.

Results¶

We validated the performance of RF-DETR on both Microsoft COCO and the RF100-VL benchmarks.

See our full benchmarks.

💻 Install¶

You can install and use rfdetr in a Python>=3.9 environment.

Installation

pip (recommended)poetryuvrye

pip install rfdetr

poetry add rfdetr

uv pip install rfdetr

For uv projects:

uv add rfdetr

rye add rfdetr

git clone (for development)

virtualenvuv

# clone repository and navigate to root directory
git clone --depth 1 -b develop https://github.com/roboflow/rf-detr.git
cd rf-detr

# setup python environment and activate it
python3 -m venv venv
source venv/bin/activate
pip install --upgrade pip

# installation
pip install -e "."

# clone repository and navigate to root directory
git clone --depth 1 -b develop https://github.com/roboflow/rf-detr.git
cd rf-detr

# setup python environment and activate it
uv venv
source .venv/bin/activate

# installation
uv pip install -r pyproject.toml -e . --all-extras

🚀 Quickstart¶

Run a Pre-Trained Model

Load and run a pre-trained RF-DETR model.

Tutorial
Train an RF-DETR Model

Learn how to train an RF-DETR model with the rfdetr Python package.

Tutorial
Deploy an RF-DETR Model

Learn how to deploy an RF-DETR model in the cloud and on your device.

Tutorial