roboflow/rf-detr

[ICLR 2026] RF-DETR is a real-time object detection and segmentation model architecture developed by Roboflow, SOTA on COCO, designed for fine-tuning.

69
/ 100
Established

This tool helps machine learning engineers and researchers build highly accurate real-time object detection and segmentation systems. It takes in visual data (images or video) and outputs precise bounding boxes and pixel-level masks for objects of interest, along with their classifications. The target user is someone developing computer vision applications who needs top performance and speed.

5,861 stars. Actively maintained with 93 commits in the last 30 days.

Use this if you are a computer vision practitioner looking to fine-tune a high-performing, real-time model for object detection or instance segmentation on your custom datasets.

Not ideal if you are looking for a pre-trained model for immediate, out-of-the-box general-purpose object detection without any custom training.

computer-vision object-detection image-segmentation real-time-analytics machine-learning-engineering
No Package No Dependents
Maintenance 22 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 21 / 25

How are scores calculated?

Stars

5,861

Forks

710

Language

Python

License

Apache-2.0

Last pushed

Mar 12, 2026

Commits (30d)

93

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/roboflow/rf-detr"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.